Baseball’s Analytics Revolution: What the Yankees’ Struggles Reveal About the Modern Game

Knowledge at Wharton

Ask about Knowledge at Wharton

Episode Description

ABOUT THE EPISODE In this episode of Wharton Moneyball, Professors Eric Bradlow, Shane Jensen, and Adi Wyner explore the latest stories through the lens of sports analytics. They discuss the Patriots’ resurgence with Drake Maye and what it means for Bill Belichick’s legacy, the Yankees’ playoff struggles and Aaron Judge’s MVP case, and how analytics are shaping modern baseball strategy. Guest Dan Szymborski from FanGraphs joins to explain how the ZiPS projection system works, what it got right and wrong this season, and how to think about playoff probabilities. The show wraps up with an incredible tennis story—a player ranked 204th in the world winning a Masters 1000 tournament—showing how data can help make sense of even the most surprising results.

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? The Moneyball podcast explains how decision-makers in the game can avoid the common mistakes and embrace the data. Episodes are recorded at the Wharton School.

#BaseballStats #Sabermetrics #BaseballAnalytics #AnalyticsInSports #SportsAnalytics ----- 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, welcome to Wharton Moneyball here on the Wharton podcast network. This is Eric Bradler, professor of
  2. marketing, statistics, and data science here at the Wharton School. And some combination of four of us today. We have
  3. Shane Jensen, professor of statistics and data science. AI Winer, professor of statistics, and data science, the three
  4. of us, and Cade Massie, some of us are here every week here on Wharton Moneyball. Uh, as always, guys, uh, we
  5. always have an open line segment, which we'll do first. As you know, we're working out some technical issues with
  6. our guest today, and we'll see if we have one, but there's a lot going on in sports that can easily take our
  7. listeners through a full hour of sports and analytics here on Morton Moneyball. So guys, why don't we start with, you
  8. know, we always like to start with what caught our eye, Shane, since you're on my left and I reread from left to right.
  9. Shane, what caught your eye in sports this week? And there could be a lot. >> Yeah, I mean, I'm sure we'll talk about
  10. baseball, so I'll wait on that. But um what caught my eye obviously is the Patriots uh and potentially having a
  11. real quarterback again, a franchise quarterback. I don't want to get too hyped up too early. We're still kind of
  12. he's only in his second season, but Drake May was absolutely incredible. Has been incredible so far this season.
  13. He's, you know, I mean, again, it's early to discuss MVP, but he would be in that conversation if we were to have
  14. that conversation at this point. Uh so I guess I'm just really excited about the
  15. Patriots. They they're looking like a playoff team. >> Yeah. What kind of credit do you give
  16. honestly Mike Vrabel very successful coach just took over as coach. Um remind me their offensive coordinator Josh
  17. McDaniel right is back as their offensive coordinator. So I hate to look I'm not taking anything away from Drake
  18. May but you've got an established head coach that's a winner. >> You've got an established offensive
  19. coordinator that's a winner. Well, winner Mike Fable. I mean, when you say winner, uh, uh, Mike Fable is a head
  20. coach. I mean, not not not his overall Tennessee record. You're not pointing to, right? You're not pointing to the
  21. fact that he's led a team to the playoffs before. >> Yeah. And I'm not trying to take
  22. anything away from him. I mean, it's obviously an impossible tell. I mean, we had like, you know, 15 seasons of Belch
  23. and Brady co-observed and people were still arguing about whether it was Brady or Belich. I don't think we're going to
  24. ever, you know, it's going to be hard to keep kind of deconvolve those, you know, certainly early on in his
  25. career. >> Look, we all believe in, you know, uncertainty, right? And I think for 15
  26. seasons for Bellich and Brady, there was obviously a very strong confound. Are you Mr. Patriot, Mr. I love both Bellich
  27. and Brady. Are is there any are you starting to at least put a small probability
  28. given what Bellich did at New England in his last few seasons, what you're now seeing at North Carolina,
  29. is there even a small part of you that's starting to give more credit to Tom Brady?
  30. >> I've always given a lot of credit to Tom Brady. I mean, that's that's a softball.
  31. Yeah. I mean, all the credit in the universe goes to Tron R. But but I mean, again, let let me stop short. I don't
  32. want to take any I I what what the Patriots did, which no other team has ever done kind of for the the length of
  33. time that they did it, was a collab, a a successful collab between one of the great the greatest quarterback of all
  34. time and one of the greatest coaches of all time. You know, I I do think Belch has done more in the post collab period
  35. to kind of take away from I guess his leg whatever, but I still think Belch's legacy is is is kind of set in stone
  36. based on what what he did with the Patriots. Would I still hire him as a head coach? No. But I mean, Audi, you
  37. want to jump in on this or on Well, you if you want to jump in on this, please do and then you'll get your own topic. I
  38. >> I do want to jump in on this. I don't think that that it's important to recognize that as a football outsider
  39. for all these years who was trying to build my knowledge base. One thing I have learned it's it's almost impossible
  40. to be successful as a coach without a great quarterback. So the fact that that Bich hasn't been so successful in the
  41. post Brady years is not necessarily that particularly damning of Bichc's talents
  42. over the years. >> How do you define success, Audi? For example, I'm just making up some names.
  43. So for example, someone won a Super Bowl with Brad Johnson. Someone won a Super Bowl with Joe Flacco. Someone won a
  44. Super I mean there are I'm not saying you get Brady's level of success. Let's not I'm not trying to compare.
  45. >> Audi said great coaches. So I think there's got to be some kind of sustained success built into that. I don't want to
  46. speak for you Audi though. >> Audi, go ahead. >> No, I mean I mean my question is is it
  47. what what is it? What is the information? Think of it from a from a posterior updating. We saw years and
  48. years of of Bellichic with Brady. Fantastic. Right. And we said, by the way, remember we saw years I you may I
  49. know you know this audi we saw years of Bellich prey too and it wasn't that great
  50. >> right? But the thing is is that there must have been more to just Brady to years and years of all that success with
  51. the with the >> Yeah. I know. And I mean, again, again, we we kind of I think um Eric, you
  52. talked I mean, or we we we think back probably most recently to their kind of end end dynasty kind of run where Brady
  53. was in god mode and in the early kind of dynastic year, you know, when they were
  54. winning very successfully in kind of Brady's very early career. >> Yeah. Brady was kind of the game
  55. manager. He did he was kind of wore what we regard as an adequate quarterback that was fine that and it was a
  56. world-class defense which Bich basically won those championships. So I I I mean it's kind of I I do think we kind of in
  57. in retrospect want to kind I I certainly give Bichc a very big part of you know that that success especially in the
  58. early years. >> The other thing to think about and this is uh something very very much a feature
  59. I think of football. the game changes. >> Um, and it does go through all kinds of
  60. shifts and non-stationarities in the way the way it's played and that in some level
  61. >> you may have been a fantastic co coach because you've mastered the skills and
  62. in the environment of a certain era and then if that if you just continue in that that you slide into your old age
  63. doing that same thing, you may not have it anymore. in which case it's not to that reflects badly on the past Bellich
  64. check but this current Bellich check is not going to do it in today's game. >> Well, first you guys have brought up two
  65. parts and I want to get to what caught your eye but I would two parts. One is I completely agree with Shane's part which
  66. is you have to give Belichc at worst if you wanted to be critical of him which there's no reason to be but if you
  67. wanted to be you have to give him credit for those first few championships where
  68. Brady was not the Tom Brady of age 30 to 40 or even 30 to 45. Um the defense was
  69. amazing. His numbers were they were good. >> He was pretty clutch in the playoffs. I
  70. mean he did win a couple but like but yes I mean I think it was definitely that was a team known for its
  71. defense. would have been, >> you know, kind of guess Russell Wilson and with Seattle type scenario, but also
  72. a your point about the non-station of the game I think is also it's another reason. It's not like I don't think
  73. Bichc has the energy or the mind or the you know it's age doesn't matter to me. It's not like I think he's lost his
  74. ability to coach people, but I think the point is is that the game has changed in
  75. a way that he would need to significantly change the style in which he coaches to be successful today. And
  76. that's just hard to do. That's hard of anybody. It's not. And also, it's hard to do for someone that won six NFL
  77. championships. And I'm not even counting the ones he won as the Giants defensive
  78. coordinator. And how many Super Bowls the man has been to? Like, I don't know, half of them in the last 35 years. I
  79. don't know what the number is, but it's some big number. But Audi, what caught your eye besides the Patriots? By the
  80. way, I watched a lot of the Patriots game sh game. Drake May is for real, but Audi, what are your thoughts?
  81. >> Okay, so of course the last time we were together, the Yankees and the Phillies
  82. were still in it and now both have been eliminated. Um so big questions going ahead um of course is particularly with
  83. the Yankees and I think the general question the analytics revolution if you if you want to condense it today and
  84. into its biggest impact factors would be the hitting revolution that has led towards the large numbers of strikeouts
  85. um and and the kind of three true outcome hitting or whatever you know >> they call that all that kind of crap
  86. >> the three two outcomes hitting, but it's essentially, it's interesting because it
  87. it it's plays off the idea that a strike out is no worse than a regular out, which conditional on an out is true.
  88. Conditional on an out. That's true. And in fact, >> no, that's true. >> Well, it's tricky because a strikeout
  89. obviously does not move the runner over and is generally a sacrifice, but it's al also not a double play. So for and
  90. from someone who watched a lot of Stanton grounders into double plays, I'm thinking why couldn't just strike out? I
  91. do think there's kind of there's an analog push against almost like against balls in play, you know, more than
  92. anything. And and and I think certain certain teams have kind of I think there's almost a counter push on that
  93. half. >> Well, so it's interesting. You look through the Yankees who with the exception of Aaron Judge, which I'll get
  94. back to in a moment, which is basically awful at the plate and massive numbers of strikeouts. No two two strikes
  95. approach. I mean, I want to pick on Trey. >> High slugging, high slugging, low batting average.
  96. >> Well, that's during the season, during during the playoffs. No, no, I mean, okay, right. Um, so very low batting
  97. average, but also very high strikeout rates. And one one of the things you noticed quite vividly in the playoffs is
  98. no two strike approach, right? So, particularly with runners on, I mean, Trent Gisham is swinging for that porch
  99. with everything he has. >> What is Vulpi like doing like >> and Vulpi is the same way, right? So,
  100. Vulpi's hits nearly, you know, about 20 home runs. >> Let me let me ask you a question. Let me
  101. ask a question. It's not like these hitters just became who they became. It's also to Shane's other point, it's
  102. not like whe you know whether it's Cashman or Boone can't see, huh, these guys bat 180 against really good
  103. pitching. Huh, maybe a good player in the playoffs. They can see the data. I I do I do think playoffs are mostly coin
  104. flip, but I'm starting to kind of deviate away from that and that I do think this kind of three true outcome it
  105. works well over very large samples. You hit a lot of bombs, etc. Uh but I do think in in the in the playoffs when
  106. you're only facing good pitchers and you have to kind of do stuff in high leverage kind of situations or or like
  107. you know the little things seem to matter or would get magnified. I don't even know.
  108. But but anyway, that that's when a sort of a like kind of hitting to contact like putting balls in play actually an
  109. ability to do that as a team is really helpful. I see Toronto and Milwaukee doing it
  110. >> for sure. >> Absolutely. I mean, one of the things we noticed about Toronto is that they were
  111. just an impossible out. They were I mean, eight, nine pitches, fouling things off and often striking out. Yes.
  112. But but tiring the pitcher, putting the ball in play. But I will say um so one of the things about so two things that
  113. caught my eye specifically in the thought around the game first of all is how unbelievably awesome Aaron Judge is.
  114. Let's get that out there. Um and I just noticed he's about the two to one favorite in the betting markets to win
  115. the MVP. Um although one of my insiders suggested that no Raleigh is going to win it and and it's you can make a
  116. strong case for either. I think um >> let me ask you a question Audi. Do they do something for the MVP like they do
  117. for the Hall of Fame or No, there is no pre-bound. It just gets announced. There
  118. is no >> well there's no prevalence. It just gets announced. That's >> just gets announced. Okay.
  119. >> And I don't think anybody tells you who their vote is. Um but I'm not sure about
  120. that. But so it's you know you know Judge had a season for the ages, right? So he'll he had a probably about a 10
  121. war. Um but just a batting line would just which is absurd. >> Feel like we talk about his MVP
  122. candidacy a lot on this show. It's not worry about it guys. He's your guys' regular season champion. Don't worry.
  123. >> No no but he had an incredible post. He had an incredible postseason and one of
  124. the things that we did see was that possibly one of the most unbelievably impressive home run I've
  125. ever seen hit 100 mph fast ball 4 in inside. >> Yeah, that was uh very few very few
  126. people could have done what he did there for sure >> just to have that kind of strength and
  127. speed to bring that bat around and keep the ball in in fair uh in such a circumstance. But here's something that
  128. I just toss up to you guys. I was thinking like what is it? I was thinking to look to see um teams that are
  129. explosive in their in their in their offense. Maybe that's a strike against them to have to just have this sort of
  130. large variance in awesome in in offense. So I thought I would rank the teams by their explosiveness which I counted as a
  131. number of either six or six runs games or more or seven run games or more. I tried both. And then I ranked every team
  132. one to 30. >> Number of times they scored that many. >> Yes. Right. So it's and and one of the
  133. things that was struck me as unbelievable this single counting statistic listed the top eight teams.
  134. I've never seen that that the top eight the eight teams that are currently in the playoffs are the eight teams that
  135. line up with the statistics with one exception depending on whether you do six or seven the Arizona Diamondbacks.
  136. Now they have a great part for hitting they were they bounce around the Detroit. I mean I'm trying to think like
  137. I mean we're not just selecting for high scoring team like I I I feel like it's kind of explosive you'd want is are you
  138. looking at the variance or just like the me like >> just a number just something triviously
  139. as simple so I would have expected that the eight teams would be in the top all in the top half but they're all in the
  140. row right here they are clear the way just to follow up on Shane's question what I think you're describing is is
  141. that let's say the average number of runs per game in the MLB is four and a half or something like that
  142. >> and so what you're describing is teams that seem to have a lot of games above
  143. the average amount of scoring and therefore by that definition they're explosive. You don't mean another No,
  144. I'm saying another way you could define explosive meaning, you know, it's how much variance is there in their in their
  145. runs per game. What sometimes they're an 11 run team, sometimes they're a three-run team that averages to seven.
  146. Like I when I when I think of the Yankees or this kind of like the bad variance that comes with like the three
  147. outcome approach is like 3-2 outcome approach. It's like you you you know you you you hit an average of like three
  148. runs a game but it's like zero and seven zero and seven zero and seven zero and seven type of thing. And the zeros
  149. really don't work well in the playoffs type. So that's so you know >> in the regular season that's not it
  150. because the Yankees have the fewest numbers of two runs or more uh or fewer and they also have the most number of
  151. six or sevens. They really had the best offense. But I was surprised that by ranking just on an off simple offensive
  152. statistic, I I thought that maybe the Padres's would be up there or you know other teams that were competitive all
  153. season long, the Cubs, they um um they just teams that were eliminated top eight.
  154. >> The Cubs were last Cubs were in it. They were in it, but the the other teams that
  155. were not um the ones that were eliminated, Cincinnati, Red Sox. Um >> you're not talking about playoff teams,
  156. though. Yeah. Yeah. Yeah. So they were >> No. I mean, I thought maybe these guys
  157. would would be Yeah, of course. But other teams, but eight right right there in order. It was just this odd oddity
  158. that that that I mean it's probably just randomness. But I've been looking for some explanation for what leads you to
  159. the playoffs. And now secondarily, what allows you to win in the playoffs and I think those are different phenomenon. Um
  160. getting to the playoffs is almost certainly unbelievable offense. Winning in the playoffs, I'm not sure. I think
  161. you have to be more Yeah, I think playoffs have to be more kind of a situ You you have to have an adaptive
  162. situational approach and I guess >> swinging for the fences every pitch is that that they're kind of trained to do
  163. over the 162 game season. That's I think what what I guess goes falls apart maybe
  164. for the Yankees specifically or for the kind of these three true outcome teams more generally. I mean the Yankees also
  165. have defensive pro ongoing defensive problems that are >> absolutely well humorous from my side
  166. but need to be taken care of. Jeez, guys. Since we're talking about baseball, we're going to do a 153015
  167. today where we've been talking open lines here uh for about 15 plus minutes. Um we do have a guest today.
  168. Unfortunately, he's only audio only, but for many of our listeners, um they're doing audio on us anyway, but we're
  169. honored to be joined by Dan Savorski today. Dan has been on the show before. Dan is a senior writer at something I
  170. look at quite a bit, which is fan graphs. He's also developer of the Zips projection system. Uh Dan, welcome back
  171. to Wharton Moneyball. >> Hey guys, how's it going? >> It's going great. Well, not for us
  172. Yankees and Phillies fans, but in general, it's going great. Um so what before we get into what's even going on
  173. today in the playoffs. Um could you just tell us Well, I was going to start with
  174. Zips, but let's forget that for a second. What do you do at Fan Graphs? cuz there's so many of our listeners
  175. that, you know, want to know continuous time win probabilities, uh, projections.
  176. What do you do? What do you specifically do at Fan Graphs? >> Well, if if you ask people on social
  177. media, I ruin baseball with math. That's that that's what I do at Fan Graphs. Uh,
  178. but but seriously, I do I I've developed the ZIPS projection system, which I've used uh for the last 20 years. Uh, I did
  179. it while I was at ESPN at Fang Graphs. That that's obviously a big part of what I do. and I just generally write
  180. coverage from an analytical standpoint. I am a statner nerd and I have been since the 80s. Uh and and that's that's
  181. pretty much what I do. >> Well, what is why don't you tell us what is the uh ZIP's projection system?
  182. >> Well, ZIPS is a set of predict predictive algorithms. Uh on broad terms, it kind of works a little like
  183. weather forecasting tools in a way. The idea of it is to get kind of an idea of where a player is and where they're
  184. going. uh the uncertainty around the future because the future is very uncertain as I can tell you from very
  185. very many wrong projections in the past. Uh if you've ever seen like a hurricane
  186. forecast where they show the little image of the hurricane on the graphic and that cone of ignorance that comes
  187. out beyond it, that's kind of the same thing that I do, just not quite as sciency as meteorology.
  188. >> So, can you give us a sense like what are some of the things the ZIPS projection system predicts? Forget
  189. what's under the hood. forget the actual algorithm itself. Like if I was to go on
  190. to fan graphs or I don't know, maybe there's a website just for the ZIP's projection system, would you be
  191. predicting the number of wins of a team? Would you be predicting who's going to win tomorrow's game? Would you be
  192. predicting who's going to win a series? What are the things that you predict with the system?
  193. >> Well, generally speaking, the the player season aligns uh also in season. There's
  194. a a simpler inseason update that can uh be run every morning uh by the Fanraft system itself. Uh so what you see if you
  195. go into fan graphs you see the projection that's kind of the median line. Uh I do when I do the team by team
  196. rundowns or I write an article uh I get you know the percentiles that things the
  197. the error the uncertainty uh because there's quite a lot of that uh uh and the idea is just to kind of peer through
  198. the fog a bit. Uh I also simulate uh teams during the season. Uh that requires a little little more work. I
  199. simulate the season a million times. uh there's a little bit of linear algebra involved when I do a generalized model
  200. of injuries to kind of capture that better than our kind of naive depth chart projections do. Uh so really if
  201. it's something I can logically model then I try to do it. Uh and you know it's it's more work than any sort of
  202. brilliance on my part. It was mostly effort and time. >> AI please. >> All right. Well, let's talk about
  203. projections and and and also reflections. Um Trent Gisham probably was the biggest surprise of of the
  204. Yankees year. Um never hit many many home runs in his history and hit 34 this year. And uh so what what did your Zips
  205. think and what does it think in the future for this guy? It's funny, my editor, one of my editors just wrote an
  206. article about the Yankees and he asked for the the the the percentile projection that Gisham's season was
  207. overall and I believe it was the 87th percentile projection. Zips has always kind of liked him quite a bit. Uh when
  208. we talk about OPS plus that's kind of what we were talking about. Uh but when you look at Zip's projection, uh I can
  209. slowly open it up and run. >> Wait, hold on a minute. You're saying that before the season that his this
  210. year's performance would have been in the 87th percentile. This is a guy probably never hit more than one home
  211. run every 50 at bats. Well, that was for OPS plus generally generally speaking, not for home runs specifically, but I am
  212. opening it and I'll tell you in a minute >> because I mean I I have to say one of
  213. the calculations that we did on our show early on, I just did a simple binomial,
  214. you know, uh p value for his rate of hitting home runs mid-season and it was in the it was in the upper tail of the
  215. of the distribution, maybe one. And >> you're saying if you used his rate as just a P and said, what's the
  216. probability under that with I'll make up 300 at bats. you have 22 home runs or whatever the number
  217. >> it was at least one in a 100 and now it's way in the 1,000s. I mean cuz the guy never hit more than just a a handful
  218. of home runs a season and I hit 34. I mean wow sorry go ahead. >> Oh but one thing to remember of course
  219. is we don't actually know the underlying probability. This is more of a beta binomial thing than a pure binomial
  220. thing. We don't actually know the underlying probability of what it is. We only know where he's been. We don't
  221. actually know how predictive that is as a as a recurring probability of that he hits a home run at any given year.
  222. >> Yeah. So then the right. So then the question is you know how much weight do you put on the you know beta
  223. distribution which is you know the prior which you could use just the raw number
  224. of at bats he has and that as a rate then you could use the current one. So the question is
  225. >> well no the the question I was doing is just looking at a a t test for the difference right is this is he different
  226. than last year? You're just using P hat. >> Yeah. Is he a different ball player and
  227. and uh and that's I mean it just he seems to be too inconsistent from what he's seen in the past to be to be to
  228. unless it would have to be very use I'm going to use my acting as host today and
  229. move away from this dreadful Yankee discussion. Dan, let me just let me just ask you a general question.
  230. It involves the Yankees sort of, but like are you surprised at all if you had used your pre-season projections? Are
  231. you surprised that the four remaining teams are the Brewers, the Dodgers, the Mariners, and the Tigers? And if yes,
  232. >> Blue Jays. >> Sorry. Blue Jays. >> Yeah, you just >> Blue Jays. >> Blue Jays.
  233. >> You had it right for the Blue Yeah. Blue Jays over the Blue Jays over the Tigers
  234. >> and the Brewers. >> It's the Blue Jays against the Mariners. >> Blue Jays. Mariners. Sorry. Yeah, Blue
  235. Jays. Sorry. The Mariners beat the Tigers. Blue Jays, Mariners, Brewers, Dodgers. Would you be surprised that
  236. those were the four teams? And if Well, certainly there's probably low probably any sim would have maybe given almost
  237. zero probability that those were the exact four teams, but what has surprised you about what we're seeing right now? I
  238. guess if you ask from a projection standpoint, Zips is moderately surprised by the exact four teams because like
  239. like most systems and most people, it picked the Dodgers uh to win the NL West, though I should know with a lot
  240. less certainty than most people did uh or projection systems. Zips only projected the Dodgers to have a 73%
  241. chance of winning the division, which is probably lower than most people would have gauged coming into the season. Uh
  242. it projected the Mariners to be second behind the Astros. uh it projected the Blue Jays fourth, but there was a lot of
  243. uncertainty in the AL East projections. There was only uh six wins separating uh
  244. the teams in in the AL East. Uh when you when you talk about the win projections,
  245. it was really really uncertain about what was going on. So, it did think the Blue Jays would bounce back, but it had
  246. him at a 40% chance of making the playoffs. It wasn't going to be the f they weren't going to be the favorite
  247. team to make the ALCS. >> So, Dan, is it fair to say the following? Um, since we're a stats show
  248. here on Wharton Moneyball here on the Wharton Podcast Network, um, since we are a stats show, I can say you're not
  249. saying that the top these four teams, the Brewers, the Dodgers, the Blue Jays, and the Mariners, you're not saying that
  250. it has a high probability of that four tuple. You're just saying it's not that surprising given the width and the
  251. uncertainty of the distribution. Yeah, I think when you look at it, no one team being where they are is surprising. Uh
  252. it's just that these weren't necessarily the most likely teams. If you asked Zips
  253. what the most likely teams uh would have been, it would have been the Houston Astros. Now, I would have been surprised
  254. because the Baltimore Orioles did not come anywhere close. Uh that is one of the larger projection misses from
  255. preseason. Uh the other one, of course, is the Atlanta Braves. Zips would have guessed that the most likely NLCS would
  256. have been Braves versus Dodgers or Phillies versus Dodgers. Uh, so nothing crazy happened, but a lot of
  257. these events were at least moderately unlikely. >> Let me ask you a question. When you
  258. think about, you know, we all try to measure success of our forecasts. um when you tend to think about the whether
  259. it's the you know the zips forecasting system or more generally what's done at fan graphs like you could say well you
  260. know uh we got the Braves wrong or we got the you know as the Braves didn't even make the playoffs so like do you
  261. think about the mag like as you're thinking about developing your system do you literally think 01 loss like you
  262. know wrong's wrong or do you think wow the Orioles wrong is really wrong >> yeah and actually just to follow up on
  263. that because I was thinking of very similar thing to Eric I think like are there particular teams or like sort of
  264. circumstances like I mean obviously the system's going to miss every year so it's got to miss something but like you
  265. know are there kind of either teams or types of teams or like is it errorprone in some way that you guys haven't been
  266. able to kind of get at yet? >> We the the the problems tend to be less bias and more accuracy. Uh accuracy of
  267. course is the hardest thing to pin down. Once you've had so much experience projecting things, you kind of can get a
  268. pretty good idea of if there's certain aspects that you're that teams are systemically being overrated or
  269. underrated. For example, if if teams with a high percentage of fast players is being uh uh consistently underrated,
  270. then you kind of have an idea and from the data uh however you choose to do that any kind of, you know,
  271. dimensionality reduction of any kind of model. Uh so there isn't really a most of these easier lowhanging fruit things
  272. have been picked which is why you see most good projection systems tend to cluster around each other pretty well
  273. with accuracy uh because the easy stuff is done and now we're kind of in very very small difference territory.
  274. >> I have a question about but Shane I saw your hand up please. Well, I guess just
  275. to kind of follow up on that cuz I mean I I totally I get what you're saying. Like I I guess what's burned my question
  276. is I I I wouldn't mind your thoughts from somebody who like looks at the data quite a bit. How are the Brewers doing
  277. what they're doing as like, you know, kind of a bottom payroll team, you know, is like do is there something
  278. kind of like that that is it how are they doing what they're doing? I it's funny that one of the things I because
  279. people ask about this with the Brewers because they're actually the team that's had the least accurate projections over
  280. the last five years uh is that if you look at the team if you look at the player projections they actually aren't
  281. systemically missing uh low on the Brewers but the team projections are. Uh, and what I deduced, uh, is that the
  282. biggest problem with projecting team standings with the Brewers, at least for Zips, I can't speak for any other
  283. projection system, which has had the same problem, is that I am doing a poor job projecting who the Brewers will
  284. actually use. Uh, for example, Zips gave a very positive projection for Joey Ortiz in his rookie year, but I wasn't
  285. convinced that they were going to go allin with him last year. Isaac Collins had a decent projection this year and I
  286. was very conservative about how many how many play appearances I gave him in the
  287. majors. It seems that I'm underrating the Braves more than the projections are. And that's kind of a problem when
  288. you're doing something like team standings because a a model such as this is not going to
  289. be able to really do a great job figuring out who will play. >> Yeah. I mean, you could have known
  290. Andrew Vaughn would go there and do whatever he's been doing as well. >> Yeah. I mean, perhaps someone can more
  291. accurately model decisions that general managers and coaches are making with with setting
  292. uh the lineups, but that's not something that's in my skill set. At least I have
  293. figured out how to cross those problems with any meaningful way. But Dan, let me
  294. ask you a question that I'm channeling my inner Shane Jensen here, even though Shane is here.
  295. >> So, this is, by the way, the odds I'm about to tell you are before last night's game. I before last night's
  296. game. So, my understanding is the if I got this right, I even got the team draw. Seattle beat Toronto last night,
  297. right? Yeah. Yes. Yeah, they beat him. Okay. But this was before last night's game.
  298. How can you explain any team Shane can jump in after you? The Dodgers were plus 120. Now, let's all be clear.
  299. There's four teams left. That essentially gives them double the probability of the average team left.
  300. How would you explain that? Like does that Dan maybe that seems normal to you? And the Brewers by the way are plus 850
  301. which gives them like you know 1/8 of probability or 1 nth of probability which is you know a quarter what they
  302. should have roughly or you know how would you explain that and do you think there's enough precision or is that just
  303. way too out of line? Well, of course, one of the fundamental things is that book makers are also seeking to maximize
  304. how much money they earn. And there's obviously going to be a great deal of correlation between the one true
  305. underlying probability, the wisdom of crowds and and where they set these things. Uh, but all I could do, I mean,
  306. is is my work. Zips at least is less sanguin about the Dodgers. >> But what does it have the Dodgers win
  307. probability at right now? Uh right now it has the Dodgers uh to just to just to get past this round. It only has the
  308. Dodgers as 5446 >> uh over that's very different. Pretend like Shane's not here. Shane would agree
  309. with that. >> So they can't have and then the Winter World Series. I don't know. Let's even
  310. give him 54 again. >> Yeah. Actually, Dan, since since you're talking about that 5446 and you've been
  311. doing this obviously for many seasons, is that kind of is that even unusual? Because, you know, you would I think
  312. most people would kind of naively look at the Dodgers Brewers as kind of a mis, you know, just based on persona like a
  313. real kind of mismatch. 5446. Is that about as much as you stretch away from 5050 for like once you get to something
  314. like the conference series or like historically after you've done this, is that like a big spread? I'm working from
  315. memory here simply because I don't have one easy to access spreadsheet but I don't believe that any of the
  316. championship series have been >> at least 6040 or or or if they are it's been very close like maybe 6139 I
  317. generally speaking I mean teams in baseball are very very close it's not like the NBA uh I can't think of the
  318. paper right now who the three analysts who analyzed best of seven series and the
  319. NBA and estimated that for MLB to have the same record of better team advancing as the NBA did that they would have to
  320. play something like best of 76 series. >> I think that was Michael Lopez who did
  321. that. >> Yes, that's it. I couldn't remember. >> He he basically said I mean he's listen
  322. because there's so much more spread in the NBA you have uh you don't need that many games to resolve who's better but
  323. because the the differences in quality in in any given game in MLB are much more compact. you need way way more more
  324. games to to >> or not even any I mean certainly the spread I mean in any given game I don't
  325. know you'd probably go above 6040 but by the time you get to the conference series
  326. >> selecting for teams where I doubt I mean I'm I'm not surprised to hear 6040 is
  327. kind of the limit of what you kind of differ from the coin flip of course I wouldn't go beyond 40 55 45
  328. >> I'm going to play the game with you Dan that I always do with Audi and Shane so
  329. Dan let me do it with Shane and Audi and and you just listen on and then I'll get
  330. your comments. So guys, we just heard 5446. I'm going to give you another piece of information and you tell me how
  331. this would move things if at all. The Brewers were the best team in baseball. They were. And they had a better wreck
  332. than the Dodgers. That move your probabilities at all? No. All right. How about this?
  333. >> Over over an even schedule? >> No. >> Okay. How about the following? The Brewers were 6 and0 this year
  334. against the Dodgers. No. Audi, is that worth anything to you? >> Yeah, I think that's worth a little bit.
  335. Tiny bit, but a little bit. I mean, the real question, Shane, is it nothing or is it is it I think it's worth a
  336. percentage point. >> Yeah, I guess. You know, again, when I think about Dan, I want to hear actually
  337. I want to hear what Dan has to say first. >> Dan, what's your thought? Is that worth
  338. any like in the I'm not saying it should. You heard Audi and Shane say if it has something, it's tiny. Is there
  339. somewhere in the zip system that has the fact that the Brewers are 6 and0 this year against the Dodgers?
  340. >> Well, directly no. But that's not necessarily right. That's just something that it's
  341. probably too small for me to model in a meaningful sense because Zips is trying to get the underlying reasons why a team
  342. might be better. But I can't say it doesn't really move it at all because Thomas Baes will will come back and come
  343. out of the grave and and punch me in the face. >> Right. Of course, it has to move it a
  344. little some way. The question is by how much. I mean, listen, I to Toronto was a
  345. beast against the Yankees all season and they beat them in the playoffs. And I I
  346. mean, sometimes that the teams are I mean, we we have a hard time putting our finger on exactly what it is, but there
  347. are different there are archetypes of players and pitchers and game styles and >> yeah, I guess
  348. >> it must matter a little bit and we have some revealed information. >> If I told you if I told you again, I'm
  349. let me frame it a different way. If I told you that team A was 6 and0 against team B, I'm not going to tell
  350. you who the teams are. Team A had a better record in the regular season than team B.
  351. Team A and team B are part of the last four teams remaining. Team A and team B are playing each
  352. other. And team A is plus 850 and team B is plus 120. You'd be like, there's some to
  353. do here. I mean, somebody would say something about something seems miscalibrated. Dan, let me give you the
  354. first comment on this on my simplistic analysis. I'm stripping off the names of the teams. I'm just giving you the data
  355. from this season. It must be because, and told us this before, if you correlate payroll with winning, it's
  356. very correlated. If you if you look at priors, obviously the Dodgers had a much stronger prior than the Brewers. Is
  357. there some other explanation I'm missing? >> Not necessarily. At least one that I
  358. think that a model can actually meaningfully address. Uh because these generalized models are are very useful,
  359. but again, we're we're talking about all models are wrong, but some models are useful. Uh and there might be something
  360. that let's say there's a particular style issue that causes the Brewers to have a better record against the
  361. Dodgers. Even if we on some level can feel that, I'm not sure we can necessarily isolate it because it's it's
  362. hard to isolate these reasons for things, it's what I always tell people coming into the season about a
  363. projection is that if a player hits 300 or I project a player to hit 300 and they actually hit 300, I still don't
  364. actually know if I was right. I still don't really know if they were truly that was the underlying probability in
  365. any given atbat was 30%. or if they were actually truly a 27% hit guy who got lucky or a 33% guy who who who showed
  366. real decline. So, you never really know the real answer in a way. I mean, there's lots of
  367. tools to to, you know, evaluate our conditional probabilities, but I don't necessarily think that
  368. it's it's in the gut. I hate to use the word gut, but there's just some little nudge here and there that models
  369. necessarily can't really pick up super well. Let me let me ask you one last question then we'll uh kind of let you
  370. go here. Um what do you see happening? There's a lot that could be going on in the off season, but like is there any
  371. player when you look at their prior on what fan graphs or the zip system might project for them and now you look at
  372. their posterior cuz the 2025 season, let's forget the playoffs just for a second. The 2025 season is over. What
  373. players if any have moved the most between your prior and posterior? like who like and possibly who may cash in on
  374. that in free agency. They could be those could be independent questions, but I'm
  375. just wondering who have you updated the most. Peter Lonzo has actually improved quite a bit when you look at the he's
  376. kind of arrested a a gradual decline. Uh but it's interesting in a way because I
  377. still don't really have a real good feel. I mean, I've talked to agents on this issue of just how much his market
  378. has recovered. Uh because last year Zips was actually pretty close on the contract he eventually got with the Mets
  379. and I was really really surprised by that. I thought, oh, Zips is going to miss by like $70 million on him. Uh and
  380. it didn't. And I was a little confused by that because there's certain things you expect to be wrong about when
  381. there's the vibes. Uh and I think he's going to get a bigger contract than he did last year. Also, there's there's no
  382. free agent compensation attached. But when he's talking a seven-year deal as he has initially, I don't know if it's
  383. recovered that much. So, I'm really fascinated uh to to see what actually happens with his contract this off
  384. season. He'll do better, but I would be really surprised if he got seven years, but I've been surprised before just
  385. maybe actually while we have you might as well get Do you have any, you know, stated projection, maybe you've made it
  386. already on how you see the uh playoffs going? >> Uh yes. Uh, Zips actually likes the
  387. Mariners before before the game. Obviously, it likes it them better now. Uh, right now Zips has them at SE as
  388. Mariners at 7228 over the Blue Jays. Uh, and then the Brew the Dodgers 5446 over
  389. the Brewers. Uh, it has depending on what the matchup is. Uh, if we just look at Blue Jays Dodgers, uh, Dodgers go up
  390. 5545. Uh, Zips actually likes the Brewers a little more than the Blue Jays. Uh, but you know, there's a lot
  391. that could go wrong for a team in that 45%. >> Well, Dan, we'd like to thank you for
  392. joining us here on Morton Moneyball. We've been talking to Dan Sorski. Dan is a senior writer at Fan Graphs, the
  393. developer of the ZIPS projection system. Dan, as always, thanks for your insights
  394. and thank you for joining us here on Wharton Moneyball. >> Always fun. Thanks for having me.
  395. >> Welcome back. Welcome back to Wharton Moneyball, the Wharton podcast network edition. Today's been an interesting
  396. show in the way we've taped it. We had 15 minutes of open segments of which Shane was going on appropriately about
  397. potential of Drake May and the New England Patriots. Ai was lamenting as he has a right to about the Yankees and the
  398. high volatility and a little bit about Trent Gisham. And then of course for the last 25 or 30 minutes or so we've had
  399. Dan Sorski on. Dan is a senior writer at Fan Graphs, the uh developer of the ZIP
  400. system. So now we're going to go back for the last 15 or 20 minutes to the open line segment here on Wharton
  401. Moneyball. Um I'll go next. So um I'm going to go to tennis because everybody knows I love tennis and um
  402. this was crazy what happened in tennis. So, just to remind all of our listeners here on Morton Moneyball, there are
  403. multiple levels of tournaments in professional te tennis. Everybody of course knows the majors, right? There's
  404. Wimbledon, US Open, French Open, Australian Open. You might as well call those 2,000 level events. The reason I'm
  405. call I'm not I'm using the number 2,000 is if you win them, you get 2,000 points.
  406. Then there's the one that's just below that of which there's about eight or nine in a year and those are called
  407. masters 10,000 events. Notice it's half but a thousand points is a lot in tennis. Then there's what's called the
  408. ATP 500 events. Then there's even 250 events. Notice I'm having there are 125 events but the major players don't
  409. really play in them. Those are the major level events. So, there are players that
  410. you guys know, many players that you guys know that have never won a Masters 1000 event. It's actually quite hard,
  411. especially now, like for 20 years, you are going to have to beat at least one if not more of the big three because
  412. they're all and I'll include Andy Murray and Stan Roinka and other guys because they're playing all playing the Masters
  413. 1000 events. And now, of course, you have Alcarz and S. And so, that's not going to be easy. and Jookovic is still
  414. playing. So, you're going to have to beat one of those guys. Well, I don't know, AI, if you saw this.
  415. I did not. Okay. What would be I know Shane has. What would be a surprising ranking? Like, how low a ranking do you
  416. think you could get for someone to win a mass 10,000? Let me just give you a piece of information.
  417. Alcarz chose not to play in this one, so that helps. He did not play in it. He had won just in China. and he won a 500
  418. event in China. Tweaked his ankle a little bit, decide not to play in this Masters 1000.
  419. Dinner played but had to retire in the second round injured. Bookovich played in the event, but like what ranking do
  420. you think someone who could win the event might be? 20, 30, 50. What do you think?
  421. >> All right. So, we actually discussed this um um not too long ago with one of our tennis guests. Okay. And my
  422. hypothesis was that um there's still a lot of dominance um outside of the top three um relative to the bottom of the a
  423. half of the top 20. But the our experts seem to think that that wasn't the case that you really get um tremendous
  424. dominance at the upper top and then it and then it and then it's it's it levels out.
  425. >> So I think probably 20 in a masters event I would say. Yeah, the number is 204.
  426. Oh my god. The guy that just won, Valentine Vashro, was ranked number 204 in the world. He
  427. beat Holgaruna, who's number six. You may have heard of him. He beat NovakJokovic. You may have heard Yeah,
  428. you may have heard of him. And then, interestingly, he played his first cousin who's ranked about 50 in the
  429. world, this guy Render Kanesh. They're literally first cousins. They're sisters or mothers. They're they played each
  430. other in the finals. Render Kesh had beaten five seated players to get to the finals, but a person ranked 204
  431. won a Mast's 1000 event. Now, let me just say he's now ranked number 40 in the world, just to show you how much a
  432. thousand will get you. And of course, I say this is life-changing, not just because of the $1.2 million he gets.
  433. That's it's twice the amount he had won in his entire career. The guy's 26. He's
  434. not 18. It's 26. It's twice the amount he had won. But he now gets into every event. So, he gets into every major. He
  435. gets into every mast's 1000. He gets into any 500 event he wants. He doesn't have to go through qualifying, which by
  436. the way, he wasn't even in qualifying for this. Just let me remind you guys for a 128 person event like this they
  437. let usually like 110 people in based on rankings and the last 18 have to play their way in. He was an alternate. He
  438. wasn't even in the list of people that could play their way in if somebody dropped out. They just had some local
  439. guy who was kind of around. >> This would be like a Would this be like an amateur winning the US Open?
  440. like like the analogy like yeah in yeah in golf I'm trying to kind of give an analogy to like you know because that's
  441. something at least we've seen amateurs come on and >> do well you know like you know I kind of
  442. feel like >> I you know I don't know if that's the right analog or if it's even more rare
  443. what just happened is even more rare than >> you should know Audi no masters events
  444. been around for about 30 40 years this is the lowest rank to ever win >> right so I I have two remarks first of
  445. all golf has more variance way more variance than tennis there's I mean you have the putting
  446. randomness which is unbelievable uh and and can make a huge difference. So um I would have imagined that amateurs have
  447. won made professional tournaments in golf every rarely but every now and last amate there
  448. >> this is the US Open though. >> Yeah there have been amateurs that have won professional golf events but I don't
  449. think if an amateur's ever well Bobby Jones in 1920s won the major >> No. Yeah. I mean I more recently I feel
  450. I feel like every every few years there's an amateur that like I mean of course the amateurs aren't going to be
  451. playing in the Masters or whatever. It's kind of you know it's the conditions for
  452. this to happen are >> the person that wins the person that wins there is an amateur championship
  453. >> that person automatically gets to play the Masters. So actually there are and
  454. there's like a couple others like I'll make it up the South American amateur champion. There are three or four people
  455. that amateurs that do get to play the Masters every year. So the real question about about this is what as a post let's
  456. be bas what's my posterior I would imagine that something we misranked him in some way.
  457. >> Yeah. Well that's that's I was going to ask like what was his trajectory before
  458. he got to 204. I don't know how far the but if this guy was just on crazy ascent
  459. or something. >> So he'd been playing for years and yet he's still I mean this doesn't make
  460. sense. Tennis is a game with lots of ladders of of levels of of quality. 204 is nothing like a 50 usually nothing
  461. like and 50 isn't like 10. So that's a lot of ladders to jump. And so >> I tell you I'm reading it right. This is
  462. look it's we I know you guys have some skepticism of AI which you should but let me just here's some data from AI. So
  463. um he's increased steadily in January of 2022 he was in the top 500 in the world.
  464. Then in January of 2024, he was in the top 200 in the world. And then by May of 2024, he had reached 116. That was the
  465. highest he had ever gotten. Then he slipped back down to 204 because of injuries.
  466. So that's it. That's his That's his entire All right. Well, at least he had been
  467. one 116 and he had a pretty strong improving trajectory which was set back by injuries. So you
  468. >> Yeah. Maybe maybe he just had like a long-term kind of crime like some kind of injury like we're looking at him post
  469. some like correction that really made a huge difference. >> I know you're doing something that a
  470. good basian would do which is that look we can't refute the data. He won. Now the question is what's an explanation
  471. for it? You've given a couple. One is the distribution of talent is really really flat. We don't really believe
  472. that but that is one possibility right? Another possibility is his true ranking is not 204 and that you know had he not
  473. been injured for part of 2025 maybe he would have been 50 or 60 in the row. Now that still would have been unlikely but
  474. not the four sigma unlikely or more we might be talking about here and this is what scientists do. You you you you come
  475. up with alternative hypotheses some of which are more plausible than another. And this one seems quite plausible. Can
  476. I can I ask a sample size question in tennis, Eric, cuz you know these ranking systems better. Like how many now
  477. tournaments in a row of like first place like first first round exits would he have to pile up? You said he's now at 40
  478. or somebody said he's like at 40 in the world now >> to get to to knock him back down to 200.
  479. How many bad performances does he like how how much was this performance worth and kind of like and sort of in the
  480. almost like I guess the inertia of the ranking system? >> Well, I'll tell you I'll give you an
  481. example right now. That's a great great question. Here's what I can tell you with certainty. He cannot go below for
  482. the next year. Let's assume that the number of points are fairly stable across the different rankings, right?
  483. So, um he pretty much can't go below about 50 in the world or 55 because that,020 points he has stays on for a
  484. year. So even if he gets zero zero zero zero zero zero and remember he's going to have he's not going to play 51 weeks
  485. but he's going to have that many weeks of opportunity to get I mean cuz he's in every tournament he wants to play. So it
  486. would take a long time now for him to drop cuz that it just happened and it stays on for a year. He's gonna he's
  487. he's changed. That's what I meant when I said he's changed his career. Like even
  488. if he has a bad year, he's probably going to stay in the top 100, which will also get him into tournaments for
  489. another year. And I hate to say it, the guy's 26, he ain't 18. You know, maybe he's not. Yeah. Everyone says, "Well,
  490. Jookovic plays till he's 38." Yeah, right. Okay. Well, Pete Sampas played till he's 31. Most players are done by
  491. 27, 28. This guy has changed the last three or four years of his career, for sure.
  492. And that's amazing. Please. >> Uh yeah. Well, I was actually one of the things we actually kind of touched on it
  493. a little bit. Um that the tennis ranking systems in particular have a lot of uh sort of built-in randomness that are due
  494. to, you know, decisions to play certain tournaments and not play them. They're not really what we would call power
  495. rankings. Um which >> they have both. They have they have the live rankings which tend to have a
  496. smaller window to them and then they have the one that I just described to Shane where the points stay on for a
  497. certain period of time. >> Right. But I'm I'm curious to know what like what >> like what an ELO system would do.
  498. >> Yeah, an ELO system or or or just someone's power ranking a regression based uh forecast that says this is how
  499. good we really think you are. And I would I would imagine that his 204 was quite probably quite different from what
  500. his power ranking or ELO based or whatever system >> like what residual would that have in
  501. like an ELO or something like that. Yeah. No, that's a great question. >> And this is actually interesting because
  502. this is one of the problems with these systems is that um it it sometimes takes into account lack of play um or a new
  503. player when he comes online, we don't know what to do with him, so we give them often a very low score. Well, here
  504. I can tell you right now again this is not I I can't see the number based before this tournament but I just went
  505. to the ATP rankings which is an ELO based system tennis abstract.com. I've talked about them before. Um it has
  506. Vashiro this guy at number 43. Now that's not very far from his other ranking but how far that move Oh, let me
  507. click on his name. Let me see if it has a time series and like like let's see how far it probably does. Hold on a
  508. second here. Well, it I if I'm reading this right, I may be reading this wrong. Fans here on
  509. Morton Moneyball. No, it had him at 204. It had him like at 200. I assume. Oh, I
  510. don't know what. Let me I I don't know what V rank is. So, maybe this is different. Maybe his rank was different
  511. before this. Oh, I see. So, it's actually changing his ranking. Oh, this might be his opponent's rank.
  512. I think these are his opponent's ranks. Yeah, no, no, sorry. That's just who he beat. Well, let me just tell you
  513. according to the ELO system in the round of 64, right? He beat the 82nd player in
  514. the world, then 17, then 23, then 31, then 11, then five, and then 54. >> That's not bad. Like what what again?
  515. Player A, player B. If one was at at 20 and one was at 200, what would you say the odds were for a single match?
  516. >> 95% at least. >> First thing that came into my mind was 99%. I would say 2% chance that the
  517. 200th player in the world beats crazy. >> There's no way he was a he was a go into
  518. that >> tour. But we we we've we've definitely the null hypothesis that he's a right
  519. >> Well, that's what I was about to say, Shane. All for our listeners out there,
  520. Audi's computing the P value under the null. Audi does not believe, as he said many times in 11 and a half years, a one
  521. in a million events happen. And so there's no way the 200th ranked player beats five, six players in the top 20.
  522. Not no way, but like with very very small odds. Therefore, you have to then say maybe the null of him being 200 is
  523. just not true. >> Yeah. I mean the question is is it should you have revised it in the middle
  524. of the tournament right I would have >> well that's well that's another thing >> certainly advance after that first round
  525. or something like that also once he beat Djokovic at that point you know I don't
  526. know I mean I understand Jookovic has good days bad days but even the bad Jookovic is still pretty darn good and
  527. so >> you know so Eric let me ask you a question because you go to a lot of tournaments and watch a lot of tennis um
  528. are you able to see with your eyes when you're watching two players that one is substantially better skilled than the
  529. other as opposed to just looking at the score. >> Yeah, it's a great question. Let me tell
  530. you what you can tell. You can tell a lot about who's hitting the ball harder because you a it's just the speed off
  531. the racket. You can also hear the difference. Like that's what I do, Audi. Sometimes I'll close my eyes and I'll
  532. try to guess which side is hitting the ball. You can see that. What you can also see and this is where I can see
  533. that Jookovic I understand now I look we can get tennis experts at some point we're going to have Paul Anakone on he
  534. would answer this as well one thing I know for a fact and the data would suggest this that Jookovic is doing
  535. worse is what they call getting out of the corners which means player a different his opponent hits the ball
  536. deep into the corner historically Jookovic would stretch hit the ball back and be back at the tea be back at the
  537. center he can't get out of the corners now because for him to stretch out far enough to get to a ball. By the way,
  538. same is true for squash. The the best players get pinned in the corners but come out of the corners fast. Jookovic
  539. just at his age now. It's movement. It's not that he can't hit the ball hard and
  540. it's not conditional on him getting there. He doesn't hit as good a shot. He does not get to the ball. He's a step
  541. slower out of the corners and that means everything in the world of tennis. The answer is yeah. That's the wonderful
  542. thing about tennis. There's things you can do that um at the live event, you can tell who's hitting it harder, you
  543. can tell who's moving better, you can tell who's able to recover from being in a defensive position more. All of that
  544. is available and motion tracking data would be able to answer all of that. Well guys, it's been kind of an hour
  545. here on Wharton Bunny Ball and the Wharton podcast network. Uh it's been a great show today with 15 minutes of open
  546. segments. Um, I don't know how much baseball we're going to be talking about now since we know the Reds
  547. >> Oh. Oh, yeah. Phillies knocked out. You're not going to be paying attention. I'm going to be talking baseball for the
  548. next couple weeks. >> All right. Well, we we'll talk some baseball, but I think we got to talk
  549. some NFL today. We got to talk about the Patriots. We didn't even get to talk about the 5-1 Bucks who absolutely beat
  550. who have the best quarterback in the NFL, Baker Mayfield. Right now, we didn't get to talk about that, but we
  551. got to talk with Dan Savorski about the zip system and projections for the season. We got talked about tennis. So,
  552. lots of great stuff. So, on behalf of myself, Eric Bradlo, on behalf of my co-host today, Shane Jensen and Aie
  553. Winer, on behalf of Kate Massie, who will be back next week. Thanks for joining us here on the Wharton Podcast
  554. Network here on Wharton Moneyball. [Music]

Brief Summary

The episode discusses NFL analytics, MLB playoffs, and a surprising tennis tournament outcome with guest Dan Savoroski from FanGraphs.

Key points

  • Drake May's Impact on the Patriots. Shane Jensen expresses excitement over Drake May's performance and potential as a franchise quarterback.
  • Tom Brady's Legacy. A debate on the contributions of Tom Brady and Bill Belichick to the Patriots' success.
  • Aaron Judge's MVP Candidacy. Discussion on Aaron Judge's impressive season and his chances of winning the MVP award.
  • Uncertainty in Projections. The discussion highlights the challenges in accurately predicting player performance and team standings.
  • ZIPS Projection System Explained. The ZIPS projection system uses predictive algorithms to forecast player performance, akin to weather forecasting.
  • Trent Gisham's Surprising Season. Gisham hit 34 home runs this year, a significant increase from his previous performance.

Episode highlights

Tension

45/ 100Some tension

Related episodes