NFL Week 3 Analytics Insights and MLB Home Run Chase & Playoff Outlook

Knowledge at Wharton

Ask about Knowledge at Wharton

Episode Description

ABOUT THE EPISODE In this episode of the Moneyball podcast from Wharton, NFL analyst Kevin Cole joins the hosts to examine what early-season models reveal beyond team records. The discussion highlights over- and underperformers, the signal in rookie and early-career quarterback efficiency after context adjustments, and the balance of coaching decisions versus roster strength. It also explores how short-yardage strategy affects expected points added (EPA) and win probability. They also talk about the wide-open college football title field, the undervalued impact of elite catcher production during a historic home run surge, and the most consequential MLB playoff races with the factors most likely to decide them.

ABOUT THE PODCAST Sports is a game of numbers. Wharton experts Eric Bradlow, Shane Jensen, Cade Massey, and Adi Wyner team up to tackle the world of sports, from current events to longstanding issues such as: What sports streaks are the most impressive? How do you rank the best players? Can athletes be compared across sports? Moneyball Highlights explains how decision-makers in the game can avoid the common mistakes and embrace the data. Episodes are recorded at the Wharton School.

#NFL #SportsAnalytics #MLB #Baseball #Football #Playoffs ----- The Wharton School of the University of Pennsylvania is committed to sharing its intellectual capital through Knowledge at Wharton, the school’s online business analysis journal.

Launched in 1999, Knowledge at Wharton offers free access to:

- Articles, podcasts, and videos highlighting Wharton faculty research and analysis of current business trends - Interviews with book authors from Wharton School Press - In-depth series of curated content like Wharton Executive Education’s Nano Tools for Leaders - A searchable database of more than 10,000 articles covering all aspects of business - A weekly newsletter that delivers Knowledge at Wharton insights directly to your inbox

Visit the Knowledge at Wharton homepage: https://whr.tn/3qUpSBE Subscribe to the Knowledge at Wharton YouTube channel: https://whr.tn/3DEAgAU

Read the full transcriptHide the full transcript
  1. Welcome to Wharton Moneyball. Welcome to a full hour of sports analytics here on
  2. the Wharton podcast network. This is Kade Massie hosting this week with two of my three longtime co-hosts. Shane
  3. Jensen is in here in the room. Audi Winer is in here in the room. Eric Bredlo will probably be here eventually.
  4. He's got some responsibilities he's trying to clean up this afternoon. As you guys know, some combination of us
  5. are here almost every week of the year. Have been for pushing 12 years now, 11 and a half years, something like that.
  6. We are recording on Tuesday afternoon as we typically do these days. Show will go
  7. up on Wednesday. We're going to go about an hour. Going to split it into halves like we usually do. Going to have a
  8. guest in here first half and then we'll do the open line segment second half. For our guest this week, we have Kevin
  9. Cole. Kevin is basically family. A frequent guest on the show, one of our favorite people to talk NFL with, one of
  10. the best analysts out there on the NFL. I think he's one of the best thinkers out there on the NFL. Always a pleasure
  11. to get some time with you, Kevin. Thanks for making time for us. >> Well, thank thanks for having me. I'm
  12. I'm pretty early in the season, so I know where I am in the pecking order. >> Well, that's right. That's high on the
  13. list. Exactly. Exactly, man. Week week week three. This is only our second NFL guest. We We're down to an hour now,
  14. Kevin. You know, we have a hard time getting all we want to do, all we need to do. We're also breaking in a new
  15. producer, so we're kind of figuring things out as we go. Baseball's been on the back burner for a couple of weeks. I
  16. don't know how I'm getting away with that, but it can't last for very long. So, we're going to squeeze a little more
  17. NFL in here while we can. You guys, Kevin Cole runs Unexpected Points Substack. That is really kind of a real
  18. delight. I want to say must readad, but really it's more like a delight for those of you interested in the NFL. Go
  19. sign up for it. Great emails come out on a regular basis. Great stuff. Unexpected
  20. points. Kevin has been in the business for a while. He ran through PFF for a little while doing his own thing lately.
  21. One of our, just to let you know, one of our favorite segments typically in the off season is Kevin doing kind of a
  22. breakdown of the front offices. He grades in in a way ranks the front offices and he's a hardcore analytics
  23. guy, man. If you're not doing if you're not doing it right, he's going to let you know and he's gonna he's going to
  24. die on that one and that's a pleasure. I think I think he makes us all a little sharper because of it. Anyway, Kevin.
  25. All right, week two in the books. That's pretty early pretty early days. So, I think one of the things that's fun early
  26. in any season, if you're an analytics person, is to see what your models are telling you that maybe the one loss
  27. records aren't telling you. You know, these are really small samples right now, but we like to think models see a
  28. little more clearly. If you if you take that perspective on the NFL so far, you know, we've got some two and0 teams,
  29. we've got some 0 and2 teams. Who is on your list of teams that are probably better than their record? And who's on
  30. your list of ah the record might be a little bit elucory the other direction? >> Well, I mean it almost feels like
  31. cheating to mention the the Kansas City Chiefs being being an 0 and2 team and and being part of that. Um although when
  32. I put up my you know power rankings as you have to do in order to get people to yell at you as much as possible on the
  33. internet. Um, you know, I always find a surprising number of people, even for the Kansas City Chiefs now, are saying,
  34. "Hey, you can't have them ranked to get over, you know, X team because because they're 0 and2 so so far this season."
  35. So, I mean, they would be the obvious one where you're looking at them and you're saying, "Okay, the record is
  36. poor. Um, the offense looks disjointed at best. the defense, you know, looks okay, I would say, but maybe not not not
  37. fantastic as it has been and kind of carried them more so last year than in previous seasons. Um, but but you have a
  38. high degree of I think they're going to figure this out and they have Patrick Mahomes and all those sorts of things
  39. together. the the real the real separation of course comes down to what does it mean for their chances to win
  40. the division, win the Super Bowl, and now the Chargers are basically a coin flip type of uh favorite to win the
  41. division and then spread out to the the rest of the 50% odds are spread out amongst the other teams in the division.
  42. So that that's going to be a problem. And they have been, you know, a team that's had the by-weeks
  43. in in this in all of these successive Super Bowl runs that they've had. Well, look, man. If you want to talk
  44. about the Chiefs being 0 and2, we're just here for you all day long. We've been dying for those guys to not make it
  45. out of the AFC. So, hey, we got time. You got time. You want to talk about that? We got time. Um, it is interesting
  46. though, you you name the Chargers being a a coin flip. I'm not sure you're probably running your own sim, but I ran
  47. a quick unabated sim using uh is it in is it's uh Inpredictable. Inpredictable. They back power rankings out of betting
  48. odds. Now, it's not like that makes it sound like it's math and it's not because you still have to decide how
  49. much weight to put the past like how how recently you want to weight things. So,
  50. there's some subjectivity to it. You can fit those things, but some art to it. Anyway, if you run those odds, market
  51. odds through the unabated sim, you do get the Chargers at 53.8% 8% 54% to win the division. And the Chiefs down there
  52. at 19 19% to win the division this time of the season. Just absolutely remarkable. If you want to roll it out,
  53. only 50/50 Chiefs 50/50 to make the playoffs and way down at two and a half% to win
  54. the Super Bowl. This is uncharted territory in the Mahomes era. >> It definitely is. and and like you um
  55. not that it's not not that life should give us a perfect uh distribution of events based on
  56. exactly how good teams are. We like luck. We like surprises. We like other things like this. But there probably is
  57. a factor when it comes to the Chiefs of being a little bit lucky. I mean, let's face it, any anytime you accomplish a
  58. lot, it's a combination of luck and skill. So, you're probably going to have a little bit more luck than others when
  59. you accomplish. It's not it's not, you know, discounting what you've done or taking credit away. It's just a natural
  60. fact that those two things are going to go together. So, teams like the Bills and the Ravens, you know, maybe if one
  61. of them can can get into the Super Bowl this year, they're probably a little bit
  62. overdue for at least having had that chance. >> Well, um you know that you you mentioned
  63. the Ravens. I'm curious where you have them in your power rankings because I was shocked at how high inpredictable
  64. has them. unpredictable. You You run their numbers in, you throw it in a sim, and you end up with the with the Ravens
  65. way up there most likely to win the Super Bowl. 22% likely to Is that win? That's probably
  66. win Super Bowl. Is that win Super Bowl? That's win Super Bowl. 22% to win the Super Bowl. The Bills are next to 14.
  67. Packers after that 11. It's head and shoulders. And the reason is because In predictable, based on market numbers,
  68. has the Ravens up top power ranking in the NFL by a full field goal. They have Ravens plus eight and a half. Next is
  69. Bills. Bills just beat the Ravens in Buffalo plus 5.7 and you got a drift under, for example,
  70. the Chiefs are down there at plus 2.9. So that's five and a half point difference between the Ravens and the
  71. Chiefs. I'm curious. I mean, that feels a little sharp to me. Um, that those differences. I'm curious where you have
  72. the Ravens and how much separation you have at the top, if any. >> Yeah, not much at all. I mean, I had the
  73. Ravens as number one coming into the season. I had them at number one after week one, despite the fact that they
  74. lost to the Bills. I had them being maybe a slightly better team in that game. I mean, things like the Derrick
  75. Henry fumble and of course the improbable comeback come into play. And again, that's a great one for people to
  76. yell at you. power rankings, you know, to keep them keep them above the Bills despite the fact that they they lost at
  77. the Bills, though. So, that's important. Um, what's what's weird though is now now they win this game against
  78. Cleveland, but they weren't great offensively in that in that game and Cleveland turned the ball over, uh, you
  79. know, as Joe Flacco has want to do. Um, so even defensively, maybe my numbers didn't say they were great. So, the
  80. Bills actually rose up a little bit above them and I have the the Bills being number one. I'll say one thing
  81. about the the the marketbased betting market based numbers over in predictable. It's kind of like Tuesday
  82. they're in a certain state Wednesday because they're using successive weeks too if they're available to try to
  83. figure out the you know the matrix to put together everything let's check back on Thursday maybe and see where those
  84. numbers at. I bet there will be less of a of a dividing line between the Ravens and everyone else and and sometimes it
  85. actually even changes up at the top once we get once we get to Wednesday and Thursday.
  86. >> Okay. Okay. That's a good that's a good warning to me and to others who might
  87. use those numbers. What can you say to us on the over on the NFC side? How do you how do you feel that looks so far?
  88. And again, we're most interested in things where the models are seeing something different from the record so
  89. far. >> Well, different from the record, I would necessarily say, but I mean the the
  90. Eagles were a team that maybe I wasn't as high as some others coming into the season because they the Super Bowl
  91. champion. they were kind of the default number one team. I had them ranked incrementally lower than the Ravens and
  92. the Bills coming into the season. Um I still have them being a little bit lower than the Packers right now. So maybe the
  93. Packers maybe I'm overrating the Packers a bit, but their numbers have been a bit
  94. stronger coming on the season. And you know, the Eagles are they get they're 2 and 0. They're the Super Bowl champions,
  95. but they haven't been impressive in necessarily in these wins, at least according to my numbers. So maybe
  96. they're a team to look at and say, you know, it's interesting to see what's going on. Um they're not as strong as
  97. maybe a Super Bowl champion 2 and 0 team that's that's that's come into the season, beaten, you know, perennial
  98. Super Bowl team uh at least appearance team like the Kansas City Chiefs already this season. Maybe I'd be fading them a
  99. little bit versus some uh market opinions. >> Staying with market, we we have the
  100. we're seeing in predictable right right on those two teams right on top of each other and the sim. So, they're
  101. positioned, you know, their relative position, their schedule, all this stuff also right on top of each other. Um, in
  102. terms of Super Bowl odds, um, any any dark courses, any interesting teams out of the out of the NFC's early on that
  103. intrigue you? >> Um, well, I mean, the Rams are interesting if, uh, if Matthew Stafford
  104. can can stay healthy. I mean, there may be a team that was outside of the top teams. Now, they had a little bit of a
  105. of a playoff run last year, so so they snuck in there. So, I think that they're kind of interesting. Um, the Commanders
  106. would be another team that I'm a lot higher on, I think, than the market. I know that Jaden Daniels had one of his
  107. very infrequent poor games last week, but it's another thing where if the defense starts to to put things together
  108. with a bunch of different pieces, there's potential to move forward. And again, I see the Eagles as being a
  109. little bit weaker than what some others may may think in that same division. Um,
  110. but this is another one where I've been higher on the Lions coming into the season. it was higher on them going into
  111. week one that looked really bad after week one, but maybe the Packers are really good and it was against the
  112. Packers. And then they came back and at least kind of reassured me that I wasn't
  113. I wasn't crazy being as high as I was on them because I value this offense and this offensive performance and they
  114. showed that against against the Bears this last week. >> Well, you know, there was a time where
  115. we thought that division might be a little tougher, you know, all the way down than it looks so far this season.
  116. Uh certainly the Packers and the Lions are legit. Um Minnesota had the great week one win, but you know, they're
  117. still breaking in a new quarterback in Chicago. Good lord. Um so, um that's that's helpful to Detroit, I suppose.
  118. You know, you mentioned Jaden Daniels. You you do a fair bit on quarterbacks and some of the main analytic stuff
  119. you've been pushing lately has been essentially like adjusting efficiency, quarterback efficiency numbers for Kevin
  120. Cole considerations. Um, I'm I'm curious. Can you say more about what you're doing, how you think it's shaping
  121. up, and then what does it tell us about the QBs? And let's do that in general, but I I'm curious about some young guys
  122. like Jaden Daniels. I mean, these are really famously hard to project early because, you know, the base rate on
  123. quarterback performance early in a career is really bad, but then every now and then someone comes in and really
  124. performs well, like Jaden Daniels did last year. Pennix has looked pretty good so far this year. I didn't see enough of
  125. that week two game to really say, but >> I'll throw in Drake May for personal considerations as well. Let's
  126. >> Okay, Drake May. Shane's got some interest in Drake May. There's lots I mean I mean what do you do when guys
  127. change teams? Do we know we got Daniel Jones apparently is a good quarterback now. So I mean let's talk tell us
  128. something about QBs. Help us understand what we're seeing. >> Yeah, it's interesting with with with
  129. rookie quarterbacks or early career quarterbacks. Um, I tend to value their performance
  130. um as being a bit more predictive than some others may say. I mean, I know it's a it's a common thing to say we we don't
  131. know a lot about them, but then if everyone thinks we don't know a lot about them, or if we can take a a bad
  132. rookie season like Caleb Williams and just, you know, toss it out the window and say we're starting fresh, then we
  133. might actually have some value flipping to the other side of saying there really
  134. is something here. And I know it's mostly directional and there are some guys who have who've had very poor
  135. starts to their careers like um Matthew Stafford for instance who was on a you know kind of a dysfunctional Lions team
  136. the 16 team that was there Jared Goff came in played maybe half a season where he looked really really poor but beyond
  137. that there there haven't been that many guys at least in recent history who've been legitimately bad as rookies who
  138. have gone on to be to be really really good. Maybe they've gone on to to be mediocre. I'm really looking at this
  139. stuff. like I'm really looking at Cam Ward and seeing what he's doing these these first few weeks and and one of the
  140. times I got people very mad at me last season of the many times was when a few weeks into the season I said you know if
  141. I was reddrafting and it's not you know it's not like a a slam dunk that I'm going to be right but if I was
  142. reddrafting I I would say you had to take Jayen Daniels over Caleb Williams even though it's only been three games
  143. just because you should assume the number one and number two picks don't have that much separating them in the
  144. first place and if Look, even three games into their careers, the distribution of uh how efficient they've
  145. been over those three games, you know, it matches up pretty well with like how well they end up doing in their career.
  146. Surprisingly well with how well they're doing in their career. >> Really? >> Yes. Yeah. So, you know, it's it's
  147. something to look at. I'm probably going to put something out after next week where maybe I'll do the same thing
  148. because I have the three the three weeks of data and I could get more people in Tennessee uh than this time mad at me
  149. for for saying what's going on with these quarters. But yeah, I mean, a lot of the rookie quarterbacks who were were
  150. outstanding, whether it's, you know, Herbert or Dak Prescott or uh Russell Wilson or, you know, these guys came
  151. right out the box looking competent and looking really good um as rookies, >> but maybe you have to look more wisely
  152. than the casual fan looks or even the longtime fan looks. You're looking at them a little more analytically. I
  153. suspect you're probably trying to condition out some things that affect performance. So when you watch Cam Ward
  154. say play two games in his NFL career so far, what are you trying to focus on or what are you trying to not focus on so
  155. that you can see his ability most clearly? >> Well, I'm I'm I'm only looking at my
  156. adjustments that I make to to efficiency. So, you know, if there's drops, I try to make an adjustment for
  157. that. If there are, you know, interception worthy throws that aren't intercepted. if there are, you know, a
  158. fumble, a sack fumble happens to be taken back 80 yards for a touchdown, that's kind of bad luck. So, so I'm
  159. making some adjustments there, but I'm also incorporating PFF grading into it. And how I like to do things often is,
  160. you know, you can model um it's a little bit more opaque. It's it's it's sometimes difficult to get to get your
  161. head around on it. I like sometimes just to visualize things. So, just if I visualize a couple of different metrics
  162. versus each other, look at the layout. Look at where these other quarterbacks have fallen even a few weeks into the
  163. season, that's when I start to get an idea of like, oh, there really is there's something here. You can you can
  164. get an idea that there's something here just simply by glancing at it. Honestly,
  165. >> what role do the surrounding cast play? It's it seems like it's inevitable that
  166. guys on worse teams look worse. I mean, just offensive line alone, receivers aren't getting open, that kind of thing.
  167. How well are you picking up on that in your numbers or or in the PFF grades? >> Well, I mean, the grading probably tries
  168. to control for that a little bit more than than my numbers. Again, we have drops and things like that or Yak um
  169. yards after catch versus expectation. So, maybe that could be something that's like a scheme or a receiver type of
  170. number that you can you can figure out. So, it it does account for some of that.
  171. But I mean, the thing is, even if you look at quarterbacks who have um emerged maybe later on in their careers, I don't
  172. know, Gino Smith would be an example of someone who had a poor surroundings and then emerged later on. I don't know if
  173. you want to say Baker Mayfield because he was kind of up or down those first few years and then, you know, the Browns
  174. probably would have would have kept hold of him if not for the fact that, you know, at the time they thought Deshun
  175. Watson was an elite quarterback who they could who they were willing to upgrade with. Um, there aren't a lot of guys. I
  176. mean, I obviously you can go back to the past like a Steve Young or something like that. So, there are some guys who
  177. go on to be like Hall of Fame level type of quarterbacks, but even so, the guys who emerge later and even Sam Darnold
  178. we're talking about here. I mean, are we really expecting that much better than average quarterback play out of these
  179. guys? Probably not. So, yeah, it matters, but I think if you're truly elite, you're going to get a hint of
  180. something almost no matter who the who the surroundings are because a rising tide lifts all boats in this type of
  181. scenario. If we're trying to predict how surroundings will affect quarterbacks, it's it's almost impossible to do. You
  182. would have said that Caleb Williams had the greatest surroundings coming into last season and Jaden Daniels had
  183. horrible surroundings and a bad offensive coordinator and everything else that people hated and then now, you
  184. know, he looks like he has much better surroundings. So, it's just really hard to predict. And when you can't predict
  185. something, I have difficulty figuring out whether it's a real phenomenon or not.
  186. >> Okay. You you said something there that I think might be a part of this conversation that we're not focusing on
  187. that matters and that is you said elite. I think maybe what you're trying to predict here is elite and what you're
  188. saying if conditional on elite you're going to see early separation and I think a lot of the confusion over
  189. quarterbacks in the last five six years has been these guys who look like utter trash to begin with and then they end up
  190. being average or just above average kind of quarterbacks like Sam Darnold for example. And that that
  191. >> Yeah, I'm sorry. And Dan Jones, maybe Dan Jones and Baker Mayfield. Baker Mayfield wasn't trash and he's not
  192. elite, but he was like, you know, third quartile and now he's first quartile or something. It's like he he moved from
  193. one part of the distribution to another. It's all still kind of confusing to the
  194. casual fans like, hold on, I thought Mayfield was any good. I thought Daniel Jones was any good. And and that is
  195. maybe that's harder to predict and you're talking more about No, I'm talking about elite pops. Elite pop. if
  196. it's going to be elite, it's going to pop early. Do is that a fair distinction? >> Yeah. No, I think that that's a very
  197. fair fair distinction. Yes. Not only the the floor, but the ceiling that we're talking about sometimes um with these
  198. quarterbacks. And I mean, it's good to it's better to be a fan of a team with an average quarterback than with a poor
  199. quarterback where you can almost write off the season. And for instance, the the Raiders, I think they one of the
  200. reasons that they signed Gino Smith is they said, "We've been so bad for so long and had such inconsistent
  201. quarterback play that, you know, getting an average NFL quarterback is probably more valuable to the Raiders than it
  202. would be would be to another team." But I think for most of the league, you really have to be targeting this elite
  203. quarterback play. That's the differentiator, especially when we talk about, you know, if we're not to get too
  204. far into it, but we get into salary cap and things like that, how much money these averageish sort of quarterbacks
  205. can make versus the elite quarterbacks. They're just such a huge differential there that I would I would just
  206. overweight that that ability, you know, finding that that top quartortile of consistently top cortile type of
  207. quarterback so far that I'm willing to ignore maybe the he hey, he could still be average. she could still be average
  208. because then we're getting two, three years into someone's career where you could have been taking more shots at
  209. someone who could be elite in the meantime. >> Okay. And just to be clear, what you
  210. mean is the the the ability is very different for relatively similar compensation when you're talking about
  211. versus, you know, middle, you know, average quarterback is shockingly similar. And so you want to play on that
  212. inefficiency. Okay. Before we leave quarterbacks, tell us some quarterbacks that you think are misunderstood or
  213. misjudged in the NFL right now in either direction based on your analysis. >> Um, let me see. So, the guys that I
  214. would say are probably misunderstood uh in a in a they're maybe a little bit better than what some people think. I
  215. mean, it's it's weird to say Dak Prescott because maybe >> he is someone who's making 60 million
  216. million dollars a year. So, so he's not exactly, you know, um >> you know, out on the streets bouncing
  217. around from team to team. But I think he is someone who has performed very well,
  218. at least by by my metrics, for a number of seasons. And even early in his career, I think he had a little bit of
  219. struggles as a secondyear player. And I don't know if it was because of his draft capital, if it was because they
  220. ran the ball so well with Ezekiel Elliott as a rookie season, but he never really got as much credit as he deserved
  221. until then. So I do think he's a very very solid quarterback, you know, with the context that he's making uh lots and
  222. lots and lots of money. Um so so that's a guy that that I would put into that bucket as maybe being a bit probably
  223. being a bit better than you might think there. Um another I would say is is Jared Goff. I mean, again, it's probably
  224. a similar type of situation where I know people look at quarterbacks doing what they're supposed to do on a play.
  225. Um, making the right throw, making it accurately, and they say, "Okay, you know, that's a you do what you're
  226. supposed to do. I'm going to move on to the next play." But if you can do that consistently and you can do that more
  227. often than someone else, even if that someone else can make a spectacular play, maybe a little bit more often than
  228. you can, uh, I just think it's relatively underweighted as a as a quarterback and it doesn't flow through.
  229. Also, I mean, I love PFF, right? I used to work there. I love everything that that they do there. But, you know, they
  230. also have a grading system that grades the throws. Basically, there's a lot of there's a lot of plays that just get a
  231. zero grading on a play-by-play basis. Now, a zero play could be, you know, could be, you know, hit the guy in
  232. stride and now he runs for another 50 yards afterwards, but it's like you did what you were supposed to do. Or a zero
  233. could be, you know, maybe you held the ball a little bit too too long, but the quarterback, but the the offensive line
  234. also allowed pressure and you got sacked, but we're going to say it was the offensive line's fault or something
  235. like that. So, you know, just doing what you need to do under certain circumstances is probably an undervalued
  236. sort of category. So, Goff has shown elite play. Like, he has actually done that, even if you don't think he's a
  237. consistent elite quarterback. So, I'm not sure for all quarterbacks, even if you put all the pieces around them, they
  238. can do that. So, he's probably someone I give a little bit more value than others.
  239. >> Okay. All right. We've been talking about quarterbacks. One of the other super important guys in the stadium that
  240. are hard to understand, hard to value are coaches. And increasingly, people are trying to get at that a little bit.
  241. Historically, we just kind of threw them into the residual and said, "Well, maybe
  242. the coaches help explain the residual." That's true, but two things. There's a lot of other things in the residual. Oh,
  243. and by the way, they're in all these numbers. They're in these stats. the coach effects are in these stats that
  244. that that that are showing up on the field in some way. So, how do you get at it? And, you know, it's, you know, we
  245. So, how do you get at it? That's a broad question. And tell us, tell us how you're thinking about NFL coaches these
  246. >> Yeah. Yeah. And I think um I could just be behind the curtain here. So, I think
  247. I I mentioned this as as a possible topic because last week the the narrative was John Harbaugh, this guy's
  248. horrible. Like, I guess I don't know. Don't ask me why. But this guy needs to go.
  249. >> Kevin, don't ask why. They've blown more twocore leads over the last however few
  250. years than anybody else. And the and the Ravens fans have like suffered as a result. So they will tell you that's why
  251. it's the late game. >> So that we so we heard this about Kyle Shanahan the year before that they that
  252. that they had this problem. So I guess what ends up happening though is again we have we have these stats, you know,
  253. that they've blown a certain win probability a certain number of times. They've blown a certain number of score
  254. lead a certain number of times. Um, you know, things things of that nature. Well, I mean, number one, it's kind of
  255. like the fourth quarter comeback stat for quarterbacks. How >> well, if you're making a lot of fourth
  256. quarter comebacks, you were also down in games a lot. So, maybe you weren't playing well going into it. So, there's
  257. like a reverse element of that. If you're losing a lot, well, that means you you were ahead. And then the second
  258. point I would make is, you know, separating these things is really really difficult. But I think what we know, at
  259. least I believe what what we know is that the players are more important than than the coaches, right? Um just in in
  260. whole. So anytime you're using a stat um and you're saying, you know, John Harbaugh has done X, you could very
  261. easily say, you know, Lamar Jackson has done X. And I think while it's not necessarily right in either
  262. circumstance, if you can pin it down, you're probably going to be right more often pinning something on on the
  263. quarterback versus the coaches. So I think what you have to do is you have to just be able to assess what we can
  264. assess. And there were some poor decisions that he made in that game. But we have to be, you know, we have to be
  265. cognizant of how big of a difference those decisions make. And normally they make, you know, a few percent win
  266. probability or something like that, which can also swing on, you know, in the course of a single play during a
  267. game, too. So, I think it's it's you you you apply blame and you say coaches should should be optimizing these sort
  268. of situations. We don't know a lot of what's going on behind the scenes, but if they've been consistently good, I'm
  269. going to give them more credit for it. Um, but we don't necessarily overweight the coach's decisions versus what's
  270. going on with the with the players on the field. >> Yeah. I mean, I I on the one hand, I
  271. agree the things that we focus on as analysts tend to be relatively small potatoes. um you know, fourth down
  272. conversions and two-point attempts and most of those are relatively small. On the other hand, the hypothesis, as you
  273. know very well and and preach is that that if they're consistently good or bad at that stuff, it probably indicates
  274. other stuff like they probably go about their business and on other domains behind the scenes poorly or well. So,
  275. it's kind of a sign. Do we see coaches who are consistently good or bad on the few clear signals we
  276. get? I know there's there's a lot of noise. Even even the coaches that we think are advanced tend to be a little
  277. inconsistent. So, do we see much regularity on either side of that? >> Yeah, I I think I think there is there
  278. is some consistency, but like you said, there's a ton of noise to it. um even the coaches that you believe may be just
  279. constitutionally conservative or aggressive and more more aggressive or more analytical if you want to call it
  280. that in some of these circumstances can be highly affected by things like a kicker missing a couple of field goal
  281. attempts in a row. Um, and and so that that can be a big big deal. And I think for for again the stuff that we don't
  282. see, there have been studies, one of them by Andrew Healey who works at at the Browns, there have been studies that
  283. have shown, you know, a winning head coach in the past is the most reasonable bet you can make as to who will be a
  284. good head coach in in the future versus good offensive coordinators and other things. So I think that's it's also hard
  285. when you're looking at what happens within a game. So, in other words, comebacks within a game, things like
  286. that. Because what what these coaches are probably doing is not really like the within the game stuff. It's
  287. organizing the staff. It's managing people. It's knowing how to lead a team. And what does that have to do with a 90%
  288. win probability? I mean, maybe it has something to do with preparation or something, but probably not. It's
  289. probably all the stuff that you're doing during the during the week that's making
  290. a huge huge difference. So, that's another reason I kind of discount maybe the in-game stuff when we're talking
  291. about head coaching, who's really like the CEO in a way of the organization. >> Yeah, it's they're so that job is so
  292. highly dimensional. I mean, that the guys who make those hires, it's so it's it's not just highly dimensional, but
  293. like unrelated dimensions, possibly negatively related dimensions. I mean, it's a it's a really hard job, but I'm
  294. not going to let you off the hook that easy. If you were a GM and you could bring in, you know, any one of NFL
  295. coaches or, you know, past No, no. One of the existing 32 NFL coaches, who would you bring?
  296. >> Okay. I can't bring Vince Lombardi back. >> Oh, you can't go You can't go to the
  297. past. Of the 32, who's top of your list? If you're the GM, >> it's a good it's a good question. Um
  298. >> because so far you've been kind of saying we can't tell other than the guys have won in the
  299. past. >> That that is a good that's a good I'm trying to think. I don't I want to make
  300. sure I don't I don't get this I don't get this wrong. So let me let me think here. I mean it's easy to say Andy Reid,
  301. but why not? you know, we have we have we have someone who's been successful at two different organizations, which
  302. again, it's it's two, but it's, you know, it's it's it's a lot more than one organization, I guess, when we have
  303. something like that. So, so I guess I would put him there, but you know, I put someone like John Harbaugh right in that
  304. same sort of sort of category. So, he would be someone who would be up there. I mean, Shawn Payeyton was someone who I
  305. said I thought he was going to be a great hire for for the Broncos. So, he's he's an interesting guy. I mean, even
  306. his brother, Jim Harbaugh, I thought that he would also be a good a good hireer. I don't know, just guys who have
  307. really had that that level of success. Um, without saying that they were completely reliant maybe upon uh high
  308. level quarterback talent or something like that, although that that doesn't hurt.
  309. >> Okay, good. Just very reasonable. Very reasonable. And it's shockingly unanalytical. Well, you've taken us to
  310. the simplest of all base rates. Just past past win record conditioning out like some stellar quarterback situation.
  311. That's it's really interesting. Okay, we're going to have to let you go, but I do want to ask one question on behalf of
  312. Audi Winer. You're a good person to ask this because you do you know we're talking about quarterbacks and coaches,
  313. but you also take a look at front offices. Why are the New York Jets so terrible?
  314. Can you just can you explain especially to our friend Audi who we recruited back
  315. into the football world but he came on this he came tethered to the Jets and now I feel guilty that we've drugg him
  316. into the football world while he's tethered to that organization. Can you explain to us the New York Jets?
  317. >> So, you know, you have this we have this like normal distribution, right? So,
  318. somebody's got to be down there, right? Somebody has to be down there. >> Kevin, that's not true persistently.
  319. They don't have to persistently be >> and in a long enough his even in a long enough timeline, you know, uh maybe
  320. we're we're maybe we're getting into, you know, monkeys pounding on keyboards and doing the complete work of
  321. Shakespeare type of timeline. Yeah. >> Um >> but yeah, I mean I I put it down to that
  322. more than anything because it's not like they haven't taken shots, I guess, at quarterback, but it's really a
  323. quarterback issue, right? Like if we're if we're saying again, I guess I'm going
  324. to default back to the players on the field versus necessarily the coaches and the GM, you also look and you say like
  325. who have they had at quarterback since uh I don't know, Joe Nameoth or something like like who have they really
  326. had who could be an elite type? >> Well, what once you start focusing on quarterback, I don't even think they're
  327. the team with the worst history. something like the B the Bears could e a Bears fan could easily argue
  328. specifically quarterback that they've had an even worse history than the Jets. I mean, the Jets, I think, are maybe
  329. worst five in in >> Am I remembering right, the great the great quarterback draft of 83
  330. was uh um Elway and ends with Montana and Kelly's in there, but then there are three others and one of them, I think,
  331. went to the Jets. So, they were in that draft. Is that right? Who was who was the Jets quarterback drafted in 83? I
  332. >> think it was Ken O'Brien, but he was the 24th overall pick I'm looking at here.
  333. So, not that >> was ahead of Marino. Marino was after that. >> Correct. Marino was after that, but you
  334. know, a lot of teams seem to mess up. I guess it just when it comes to the Jets,
  335. we have, you know, we have, you know, Zach Wilson number number two pick. Sam Darnold number three pick.
  336. >> Uh Mark Sanchez number five pick over >> Chad Pennington was pretty good. Remember Chad Pennington?
  337. >> Chad Pennington was >> I say he was decent. I mean I mean again during the Brady years it was kind of a
  338. tough slog you know that was even >> can I get some context here how bad are the Jets in sort of standard deviations
  339. are they >> two what period of time >> or five >> no not five not five >> no so where are they how I mean it's it
  340. feels horrible but I'm biased so what what is in our in our adult life let's just let's call us an average of I don't
  341. know I don't want to impugn Kevin too much let's call us an average of 50. Me and Audi gonna take you guys down. Um,
  342. so the past 30 years, the past 30 years, >> what do you think the Jets are win loss
  343. record versus the league? Minus one and a half standard deviations. >> I feel like they haven't even
  344. necessarily I mean I mean I I guess maybe bottom third team, but I wouldn't put them much below that when there's
  345. teams like the Browns and stuff like that. What they've done over the last 30 years.
  346. >> Yeah. Yeah. the Browns, the Jags probably are another >> I mean look at the Lions up until a few
  347. years ago and I mean that's the hope you have. I mean you know any any team can have a 40 50 86 year run of terribleness
  348. right and then turn it around. So >> you're right the recent history you know it's like the last 15 years your one
  349. winning season was when Ryan Fitzpatrick had a had a great season. So, you know,
  350. yes, you're right. The last 30 years, if you weight them all equally, but if you're waiting the last 15 years a
  351. little bit more heavily than the prior 15 years, that's when the Jets start to look.
  352. >> That's true. And that's what that's the Wharton Moneyball relevant window. We
  353. could just say last, you know, this is 12 seasons that we've had Wharton Moneyball that we drug Audi into this
  354. world and he's just been beat about the B. >> They haven't had a winning record in all
  355. that time. >> Yeah. Okay. That's we're going to start working with the with the with the
  356. Wharton Moneyball era. I think that'll quickly become a thing, the Wharton Moneyball era. Um, all right, Kevin, we
  357. should let you go. Thanks for making time for us. Always, always enjoy the conversation.
  358. >> Thanks for having me. >> Kevin Cole, pick him up on Unexpected Points. Says, "Subsack Unexpected Points
  359. is fantastic reading if you're interested in analytics and the NFL and you can catch him here periodically and
  360. many other places." Kevin Cole, welcome back to Wharton Moneyball. Welcome back to the second half of this week's show.
  361. An open line segment with Kade Massie, Audi Winer, and Shane Jensen. Some chance our buddy Eric Bradloow slides in
  362. here after he wraps up some business on campus, but probably not. We'll probably
  363. have to wait another week to catch Eric. Dion Simpkins, running the show behind the scenes. It just struck me, guys.
  364. Dion counts us in. You know what Dion does in his spare time? He plays drums. He counts musicians in all the time. I
  365. didn't realize after 11 enough years that he's doing the same thing with us that he does with his bands. Counting us
  366. in. Dion Simpkins. We depend on him. Big Eagles fan. Dion. Big Eagles fan. He's had a good run lately. Audi at the break
  367. was musing about the Eagles. We're just off the line with Kevin Cole. We talked NFL for half an hour, first half of the
  368. show, and Audi is musing about the Eagles. What do you got, Odd? What are you worried about?
  369. >> Yeah. So I I just um recalling a couple years back actually maybe maybe many years back I built a model to predict
  370. next year's performance given the past that was incredibly simple and it essentially took two elements last
  371. year's performance and the average of the three years and it was a really good predictor. It is about almost as good as
  372. the odds which the current odds which are which are which are very updated in their in their data and should be much
  373. better because they have information. And so I asked the question um given our discussion with with with Kevin that the
  374. Eagles don't seem to be one of the top three or even five teams um in the league right now. And what happened to
  375. the Eagles over the break that my simple model would have missed? They've already
  376. now won two games. And maybe it's because they didn't look too impressive in the two games. Is that what it is? Or
  377. is it that something happened over the offense? So, can I do one one quick bit on your model before we go to the
  378. Eagles? So, in terms of just ordering, I can imagine that that would do well, but
  379. in terms of >> variance explained, it can't do that well, right? Is is my memory of the
  380. persistence of records is that there's a fair bit of regression to the mean. Like
  381. maybe maybe a third of the record persists. >> Well, yeah. You know, no, there's a
  382. coefficient that cuts them down. It's just the inputs to the model were just those two.
  383. >> Yeah. Yeah. I don't know. I I mean I don't know if I mean Kate you could maybe correct me if I'm wrong. I don't
  384. know if there's that many power rankings that have the Eagles out of the top five
  385. in the league. I mean I think >> but not one >> one Well yes one could right several I
  386. mean but you know I I think probably the teams the two teams at the top of last year's power rankings I bet weren't in
  387. the Super Bowl. It's probably the Bills and the Ravens, you know. So I I I so you know I think maybe
  388. >> I I I think there's some top teams out there that obviously you know weren't in
  389. the Super Bowl last year. >> I think what you're saying they weren't the best team last year by the
  390. >> by the by the power. They might well not have been. I think the Ravens were probably top.
  391. >> I think that's right. But, you know, I think some people have talked about there may be ways that they play that
  392. are outside of good models. >> Like for example, >> certainly I I think >> the tush push people the tush push is a
  393. a quality of their team that's just not picked up and that might be >> substantial that it might be enough. It
  394. might be significant. It might be enough to to make >> I actually we were actually talking
  395. about that at lunch over our analytics meeting. the fourth down play where you can increase your your conversion
  396. probability as significantly as the Eagles can do it with the tush push from probably low 70s to mid 90s at a with a
  397. play that has that much leverage. I mean, think about in football a fourth down play has incredible amount of
  398. value. That's just I think that's just not captured. >> Then it's not even it's more than that
  399. because I mean basically if they can get like seven yards on first down it's a guaranteed it's guaranteed they will
  400. >> right because they've got one or two like you I mean maybe you stop the push push once you cannot stop it two three
  401. times in a row. It's just not doable. So it's kind of a cheat code. >> Okay. So that I'd be interesting to
  402. know. We I we it'd be fun to have a modeler who's tried to work on that and ask how much difference does that make?
  403. Like what are we missing? What are the traditional models missing by not having >> the the other thing with the Eagles that
  404. I think gives them kind of has given them at least observationally a positive residual the last few years is they have
  405. a defense that always shows up against good QBs. And I think that's kind of exceptional. I saw somewhere, I think
  406. maybe it was on Reddit, that since the start of the 2023 season, the Eagles are 9 and0 against teams with like a top
  407. tier QB. The the tiers here are defined by the athletic. I'll tell you who they are. Allen, Burrow, Jackson, Mahomes,
  408. Stafford. >> Yeah. >> And against those teams, Rams, Chiefs, Bills, Bengals, Ravens, the Eagles have
  409. are nine and0 since the start of the 2023 season. >> That's absurd. So sort of exceptional
  410. performance against top talent. >> Can I throw out another possibility? This is again related to the tush push.
  411. Isn't the EPA per play becoming a dominant measure? And EPA per play is expected points added per play. And if
  412. you have if you can extend your your plays because you can because you have the tush push to convert short ones,
  413. you'll have a lower average EPA per play because you have more plays in your possessions. And that might be something
  414. that is that is not being accounted for. Um or is EPA because I'm hearing EPA per
  415. play everywhere these days. And I always found that a little bit odd because shouldn't it be EPA per possession? But
  416. that's not the right thing. That's >> Yeah. I I I also don't I don't work like as much with EPA as kind of like a a
  417. modeler would, but like uh I would assume that like you'd want to first of all cut out like EPA per play on passing
  418. versus rushing like you you know you you I I think there's certain like kind of ways of subsetting the data where you
  419. that would make it a lot easier. >> So So Kade, what is your way in this APA per play? What is your you know do you
  420. see the problem as I see >> I thought you were going somewhere that you that you didn't go and I turned my
  421. attention somewhere else for a moment and I realized that you had gone someplace unexpected. So restate your
  422. issue with EPA. >> My EPA EPA So EPA per play is expected points added on a per play basis. That
  423. seems to favor um teams that get it done quickly as opposed to teams that ek out a little
  424. extra play expected value over many plays because then your EPA per play is going to be lower as opposed to with big
  425. big plays and moving the ball quickly down the field. But it's expected points. So that doesn't it automatically
  426. adjust for fourth down? Like a fourth and two being more like two yards on a fourth down would be actually more
  427. highly valued than for two yards on a first. Yeah. You know, so it almost maybe rewards for getting stretching it
  428. out into those fourth downs. >> So you don't you don't So the issue is it's if you if you get a little bit of
  429. value on a lot of plays, you get you to the same place as getting a bunch of values on a fewer number of plays. So I
  430. so it's a interesting question and I don't I don't have the goods on it but a couple of observations. I I think
  431. there's I suspect that that if you're just barely ekking it out that that that there's information in
  432. that and that and and the chance of stringing a bunch of those together breaks down. Um on the other the other
  433. end is is true as well. Like big explosive plays are not very predictive. they they matter a lot in the outcome of
  434. the game, but they don't tend to persist. But there I think where I think what EPA captures well is kind of that
  435. middle ground where you're in general you're you know you're getting chunks you're just getting chunks at better
  436. rates than other people. The I think your argument would break down Audi because I I think if you're just ekking
  437. it out it's like teams that barely win games there's not this special thing about being able to win. They're just
  438. generally >> or you're you're almost talking about like what would be the strategic value
  439. in a team basically just tush pushing the whole time like what happened if that was the only thing they did?
  440. >> Yeah. >> And how they kind of crawl could you crawl down the field basically like
  441. that? >> What I'm conjecturing is is there a team for which EPA per play doesn't capture
  442. their value? >> That's essentially the question. And what would it look like? And what I
  443. would argue is it's a team that has this ability to to get a couple yards when they need it on a really regular basis.
  444. >> Yeah. >> And really what you're talking about there I think is well what's not being
  445. captured by the EPA the expected value is how much variation like at the team level how if a team really vary you know
  446. there is a team that could consistently get that expectation every time would be
  447. valuable. Exactly. Low variance in the in in the very low E small EPA but low low variance. the question. Okay. Well,
  448. now you've now you can go look at data to see whether there is such a thing, whether that is a trait of a team,
  449. whether that's a consistent thing. >> Sure. I mean, we must observe it sometimes, but is it just by chance or
  450. is it persist? Is is that it would be tied to, you know, the style of play or really good quarterbacking maybe? Um,
  451. but I think it's an interesting question, guys. We should talk about something besides NFL. We've been doing
  452. NFL for pushing threequarters of an hour now. So, of course, let's talk about college football. Um,
  453. >> come on. Get a little baseball. Yeah. >> Was a player of the week in the American
  454. League last week. Hold on. Give me Give me >> Let me just before we transition before
  455. we transition to baseball, let's What do we have to say about football, college football? Um,
  456. >> I don't know, just as a summary, I I looked at the title odd. So, we're we're
  457. three weeks into the college football season, four if you count week zero, but mostly three weeks in. And the title
  458. odds, the the remarkable thing about the season so far is that it still seems quite open. So the title odds are quite
  459. flat at the top still, but I was curious where they were. And what the most interesting thing to me is that the top
  460. three teams, the shortest odds right now, Tuesday afternoon, are Ohio State, Penn State, and Oregon, who are all
  461. members of the Big 10, which is a little bit surprising. And I and I think that's
  462. because there's not much else going on in the Big 10. like we can be pretty dang sure those three teams are going to
  463. make the playoffs. More confident than any three teams in the SEC and I think that's where their advantage is coming
  464. from. But it's it's it's interesting and kind of unique that you have Ohio State,
  465. Penn State, Oregon at the top of the the odds board. >> And it's kind of interesting because
  466. normally you would sort of say like you know like >> because because the foot because the
  467. playoffs are more open now and their path doesn't necessarily exclude each other. they don't they're not competing
  468. for the same kind of allocation. Before like when we were talking about kind of the football odds, I was surprised that
  469. the Bills and Ravens were like the top two just because they kind of they have to go through each other basically.
  470. >> That's exactly right. So, and so I I think that's I think that's that that and the and the different the difference
  471. in the Big 10 versus the Big 12. If you look at the not the Big 12, the SEC SEC has six of the top 10. Eight of the top
  472. 14. I mean, that's just absurd. Eight of the top 14. It's just incredible. ACC has three teams in the top 20. I'm
  473. talking about top 20 shortest odds to win the championship. But the highest one is Miami. Miami at eighth. And then
  474. the Big 12, the fourth of the four divisions, really the power four divisions have two teams and they are
  475. 18th and 20th. So Texas Tech, get your guns up. Red Raiders, highest rated odds team for the championship in
  476. the Big 12 at 18th and Utah just a little behind them. By the way, that's a good promotion. That's a game this
  477. weekend. Texas Tech at Utah. So, at the top of the Big 12 and if you can if you if you're interested enough in college
  478. football to be interested in the Big 12. Good Big 12 match up there. And a second
  479. one, ASU, Arizona State going into Baylor. So, a couple of big Big 12 matchups this weekend.
  480. >> How much title probabilities left over by the time you get down to the 18th or
  481. 19th ranked team by probability? >> Not enough to worry about. How's that? I think three weeks in.
  482. >> Right. Right. Don't don't buy be buying. >> Yeah. Right. And like even even when
  483. you're talking about winning the whole damn thing, Shane, I think we can ignore after 20 even before the season begins.
  484. We the guys are going to make the playoffs that come out of the top 20, but they're not going to win the thing
  485. coming out of the top 20. I, you know, it'd be nice to know how much how many teams you need pretty flat right now.
  486. How many do you need to capture 50% of the probability? Audi could run some odds conversions for us real quick, but
  487. you're gonna need, you know, you're gonna need four or five teams, which is unbelievable in college football that
  488. you would need that much to capture 50% of the probability. It's just not been that flat in a long time. Um, all right.
  489. The we have a lot of good games, but there aren't like monster games on the college football side. NFL is a lame
  490. lame schedule this week. Let me just say Detroit of Baltimore Monday night is the
  491. the headliner I believe. But even the Lions are six and a half point dogs going into Baltimore. Okay, that's
  492. football. Football. What about baseball, guys? What do you got? We got the home run race, I think, is the most
  493. interesting thing right now. No. Or at least the number >> I'm looking at I'm looking at two
  494. forecasts I made many months ago and as as they're starting to resolve, they had
  495. huge uncertainty anyway. So, it's not exactly as if if losing is great, but I did predict Cal Riley at 57.
  496. >> Um, so, um, >> he slowed down a bit. He's still I think he's still stuck at 54.
  497. >> And I had and I predicted Judge at around 323 months and months ago is his final
  498. batting average. Um, so Judge has been hot. So, he's got 48 homers. I think he hit five in six games or something. um
  499. after being being on the disabled list. So that does it does ask the question who's going to win the MVP in the in the
  500. American League >> um judge. >> You think it's an automatic judge? >> Yeah. I mean
  501. >> if I had to actually >> all the all the rate numbers argue guys before you go too fast pass the
  502. home run because y'all go in MVP and now I need more dimensions than I'm paying attention to. Home run's real simple.
  503. It's real simple for like simple distant baseball fans like me. We see some guys
  504. here. We see some numbers. They're all around 50, which is amazing. Do you have anything else to say here? I mean, I I
  505. remember y'all being y'all poo pooing that Judge wasn't going to win the thing before he got injured. And now that he's
  506. back, he's making a good run at it. But if we get all five of these guys over 50, how how unprecedented is that? That
  507. is unprecedented. >> Well, I mean, yeah, if if we were to get five, uh, that it would be the first
  508. time in baseball history that that had happened a couple times. >> Suarez is at 45. We only have two weeks
  509. left. That might not happen. What if you say four? >> So, the only two previous times that
  510. we've had four in history were both kind of steroid era. I think it was 99 and 2001. I could list off the guys
  511. involved, but it's, you know, Bonds, Magguire, those those dudes. Okay. >> Uh um so this would be certainly the
  512. first non steroid era kind of instance even of four. >> And the fun, you know, >> the fun thing here is that no, nobody
  513. would be surprised at Schwarber Otani Judge being at the top. I mean that's those are like those are like straight
  514. out of the you know top of the queue but then >> perennial perennial home run hitters
  515. >> but then Raleigh is actually leading the list and it's just I think it's been one
  516. of the great stories. Um I I I'd love to see a good backstory on that whole thing. Um
  517. >> see the the reason why I'm I mean so what makes Judge different than the others is he has an OPS of 1150 or
  518. something. His b his batting average is almost 100 points higher than Raleigh. his on base percentage is 100 points
  519. higher and if you watch you he's just always on base and uh it just >> he's one of the greatest regular season
  520. players we've ever seen. >> Yeah, thank you very much. He just comes out and got him.
  521. >> And there you have it. There you have it. >> Um but uh but so it's hard not to look
  522. the other way. On the other hand, Raleigh's doing it a catcher. And that is, I think, the position that is most
  523. poorly valued because it's the only position where you can have an enormous gap between number one and number two on the
  524. offensive side. You just don't see that at first base or the outfield or even shortstop. You third, you just don't see
  525. one guy having that good a season relative to who's two. which means it's just an automatic competitive advantage
  526. that doesn't exist in any other position. And I don't know how people properly evaluate that. I'm not saying
  527. >> I guess the way the way I think about which I think is mostly just paraphrasing you is that I think it's
  528. catcher is such a difficult position fielding wise both grueling in terms of like the stamina involved as well as the
  529. g the various intricacies of both of both throwing as well as game calling. there's so it's the closest thing to
  530. kind of an onfield coach that you have. So there's all kinds of that skill set and then to also have that and be a
  531. great hitter. It's just I think more rare to see that kind of combination of skills involved. And that's why, you
  532. know, when you look at Hall of Fame hitters, you know, and and you're comparing across positions, obviously
  533. the catchers don't necessarily look on paper as well because we're kind of baking that into our evaluation, I
  534. think. And so when somebody like Cal Rally stands out or Joe Mau a few years Joe Mau won MVP I think as a catcher um
  535. you know sometime when they kind of do go off like this I do I agree with Audi it really stands out in in a way that
  536. kind of you know >> Yeah sort of norm for Judge doing what he does every >> Yeah. So body talk about how the how we
  537. so that all makes a lot of sense to me and you're saying it's not capt like as as good as baseball is capturing value
  538. they're missing that >> they're they're missing that because they sort of they they so everything is
  539. additive right so you you don't recognize that you have to have a catcher you can't you got to do it right
  540. so there is no way if you have Cal Rally on your team doing what he does no other
  541. team every team is at a big deficit automatically out out the gate relative to the team that
  542. >> shouldn't that be captured in war if it's war is position specific >> if they could do kind of
  543. yeah I I mean so I mean you're right though Kate that at least you could try and defile define build that into the
  544. value like what the concept of a replacement player is and but I don't think that you know because of the kind
  545. of attrition that happens as catcher and stuff like that the selection bias against catchers means I think it's very
  546. hard to define that replacement value and Maybe I guess Audi's arguing and I would support him that you should sort
  547. of almost tweak like that the concept of replacement player for catcher should be
  548. kind of brought downwards >> because it's such a rare skill set to be competent at all that
  549. >> so it is it is down but here's my here's the observation that I make about war
  550. and how it deals with position and position of catcher in particular. It's invariant to the distribution of talent
  551. above the replacement. Think about that for a minute. So if you have if you think of the replacement
  552. line, there's always someone available right at or below replacement. That's the assumption. And that there's someone
  553. available above the replacement, but doesn't ask the question, what does it look like? So imagine a just the
  554. placement level is zero. Imagine there's one guy at 10 and another at nine, another eight, another seven, another
  555. six, another five, all the way down, a big pile of three, and two and one. That may be one way it could be distributed.
  556. And now imagine a a different position. There's one guy at 10 and the next are all at one and below.
  557. >> Yeah. No, Audi's made a great point because you're >> telling me that that 10 the 10 in each
  558. group are the same and >> and that's something you can't correct with just redefining the replacement
  559. player because you're talking again about sort of the a different distribution like we should value
  560. somebody who's a fivew player at catcher >> well the replace more than somebody who's a five war player in the outfield
  561. because there's just not that many five war players in major league baseball >> right so go with me for a moment like
  562. imagine you're running a fantasy draft Right. And you got there's one catcher at 10 and all the rest are one and
  563. lower. And at center field or in the outfield there's one at 10, one at nine, and one at eight and a seven, a six,
  564. whatever. There's lots, right? >> Who you picking first? If you got the first pick in the trash,
  565. >> I don't you know, it's the 10 at catcher because you just got now nine points
  566. over all your competitors that they're never finding. >> So, it should make that somehow more
  567. valuable. And that 10. Now, you could say if the 10 is already in there, if they've put the 10, but it's not because
  568. I know that the war calculation is relative to replacement value. So, they don't look at that distribution above
  569. the replacement value to find out to figure out your war. And so, I want I can imagine doing that. So, how does it
  570. relate to the MVP race? It it my gut says that if if Judge doesn't have a huge offensive advantage, which right
  571. now he does, but if it's not huge, I'm awarding it to the catcher. Yeah, that makes a lot of sense.
  572. >> Yeah, it's kind of like, you know, if if it was a DH that was kind of running
  573. away like, you know, I think we kind of like like somebody like Schwarber maybe gets dis, you know, naturally
  574. downweighted relative to somebody who's fielding more complex positions. So, I guess he's going up against DH
  575. pitches. >> Guys, we are down to the last two weeks. Are there any races to We haven't talked
  576. about races in a while and yet they're we're we're at it now. I mean, >> oh my god. Well, it's Yankees Red Sox
  577. right now all over again. >> I mean, they're kind of lined up to play each other in the wild card round if
  578. which I I I personally would >> Well, there's a home field advantage that's really up in the air.
  579. >> Right now, the Red Sox have an advantage in in head-to-head play. So, the Yankees
  580. actually have to come out ahead, not tied in order to uh in order to get home field. Um, that's a big race.
  581. >> By the way, the Brewers have have sealed it up. And h how the hell they did it at
  582. this point is just remarkable without any stars. Talk about we asked the question what what would you rather have
  583. in your distribution of talent lots of lots of twos and threes or or is it an extreme and you know how does that work
  584. and talk about a recipe for for winning that no one really predicted at this level I think.
  585. Yeah, certainly. I mean, you know, somebody would I if if somebody asked who's going to come out of the central,
  586. I might have picked the Brewers, but I would not have had them as the best team in in baseball.
  587. >> Well, to make explain, one of the things y'all are saying is that the Brewers and
  588. Phillies are vying for the National League home field for the for the for the duration, right there. That's the
  589. That's a real consequential battle that it's in. >> It is. >> Okay. So, that's that's interesting. Um,
  590. looks like there's some interesting wild card stuff going on. >> Well, I mean, I think the the AL West
  591. race I think is the interesting one in ter, you know, as well because I mean, >> well, I know. Well, yeah. Oh, I'm sorry.
  592. AOS AO AO West I think is the more interesting one just because, you know, we've got the Astros Mariners and I mean
  593. the Mariners in addition to that being like, you know, a half game or game separation between the two, I think this
  594. is the Mariners have the longest postseason drought streak I think in >> like professional
  595. >> in American sports. So that's right. That's right. >> In American In American sports. So, um,
  596. yeah. I mean, obviously that I I kind of feel like you we got, you know, I mean,
  597. I'm not sure how many casual fans would be out there rooting for the Astros anyway, but I I feel like the public
  598. should be really on the Mariners for this one. >> Yeah, for sure. For sure. Okay. Well,
  599. there's a lot of interesting stuff going on. You're talking about So, the Jays somehow clinch the not clinch, but
  600. they're going to win the East. So, it's about Yankees Red Sox for the wild card there. We're talking about the M and
  601. Astros, right, Shane? That's a great team to pull for. >> Yeah, the Rangers aren't far back. So, I
  602. mean, you know, the you know, the Mariners and Astros have uh have still still got to keep it going as far as,
  603. you know, one the loser of that race could still miss the playoffs. It's not like they're just competing for the
  604. division. >> Right. Right. Right. Okay. And then flipping over to the NL, I mean, the
  605. Phillies have made the East a sleeper, but they are >> Well, the Mets have really made the East
  606. a sleeper. The Mets are really something we should mention. They were the top team top team in the NL
  607. >> at mid Jul mid June I think up until mid June and now they they're barely holding
  608. on to that like last wild card spot. >> Yeah. Bar it's an incredible collapse. >> Barely barely barely. I could they could
  609. this could easily go away especially if they continue what they've been doing. >> Um but then that that that NL West is
  610. that's a great rivalry in baseball these days. this Dodgers Padres's thing. The Dodgers do have a two-game lead right
  611. now. Um, which is pretty healthy with only two weeks left, but that that's the Padres's haven't won many of those
  612. divisions, right? That's that's a that's been a hard thing for them to get over in recent past. So, that's a fun one.
  613. All right. There's a lot of good things for us to pay attention to as baseball comes down to the wire here. Um, fellas,
  614. why don't we leave it there? Why don't we call that a day for the whole crew here, Audi Winer and Shane Jensen who've
  615. been in here from the beginning for Eric Bradlo and Absentia, our fourth co-host.
  616. This has been Kade Massie. Big thanks to Dion Simpkins. Big thanks to D Patel. Big thanks to Marissa Raina. That team
  617. keeps us going. And thanks to you guys for listening. Come back and join us next time. Between now and then, enjoy
  618. your sports. [Music]

Brief Summary

Kevin Cole discusses NFL analytics, team performances, and quarterback evaluations on Wharton Moneyball.

Key points

  • Welcome to Wharton Moneyball. Kade Massie introduces the show and its hosts, setting the stage for sports analytics discussions.
  • Kevin Cole Joins the Show. Kevin Cole, a favorite guest and NFL analyst, discusses early season analytics.
  • Chiefs' Surprising Start. The conversation shifts to the Kansas City Chiefs' unexpected 0-2 record and its implications.
  • Quarterback Predictions. The discussion revolves around predicting quarterback success based on their surroundings.
  • Misunderstood Quarterbacks. Dak Prescott and Jared Goff are discussed as potentially underrated quarterbacks.
  • Understanding Coaches. The difficulty in valuing coaches and their impact on the game is explored.

Episode highlights

Tension

40/ 100Low tension

Related episodes