Assistant Teaching Professor, Director of Carnegie Mellon #SportsAnalytics Center https://xitter.cf/t.co/bH22Ka8HwC @CMU_StatDS
Pittsburgh, PA
Joined March 2013
- Tweets27.4K
- Following1.3K
- Followers7.2K
- Likes26.5K
Pinned Tweet
It's a privilege and honor to be able to work w/ @HeismanTrophy - I consider it to be the most prestigious award in American sports, and I'm beyond excited for #CMSAC to play a role in its rich history! #Heisman #CollegeFootball #sportsanalytics
CMU x @HeismanTrophy 🏈📊🏆
Heisman has partnered with Carnegie Mellon and the Sports Analytics Center to develop metrics and tools to make it easier for fans to access and evaluate performance data for all positions.
'The Big Bet' podcast w/ @dannyfunt is tremendous, I strongly recommend listening to it (looking forward to the final episode) - the fact that this podcast as well as his awesome book 'Everybody Loses' is not being discussed says a lot about sports media podcasts.apple.com/nz/podcas…
Ron Yurko retweeted
@RyanBrill_ and I demonstrated that uncertainties aren't not small. Even in a simplified game there is not enough independent data to make confident decisions in many situations. degruyterbrill.com/document/…
I think Brian is great, but I'm sorry - this is not a good thread, I honestly think I'm going to assign students in my #sportsanalytics class this spring to critique it! I look forward to talking to Brian about this in a few weeks at FAST though! csds.tamu.edu/football-analy…
Ron Yurko retweeted
This was a game changing paper on how we approach 4th downs on the show.
I think the biggest takeaway is if teams don't practice and gameplan 4th down differently than 3rd down and short yardage we have a problem
Ron Yurko retweeted
Biannual reminder: when people say "probabilities" they often should be saying "estimated probabilities". Estimates can be (often are) far from rock solid: complex feature space & sample size is smaller than you'd think --> we're uncertain about the estimates we've produced
Two things can be true:
1) Shanahan has historically been too conservative in 4th down situations
2) Analytics twitter is overconfident in their win probability estimates
REPORTING UNCERTAINTY MATTERS arxiv.org/abs/2311.03490
Kyle Shanahan says a 51% analytics recommendation won’t convince him to go for it because he gets one shot, not an hour at a blackjack table
“I don’t want to hear about it if it’s 51% of the time we should go for it.”
“As I always say, I’m not sitting here playing Blackjack for an hour with no one else at the table, and eventually I’ll get 51% of it. Like, I have this one try, and I have a feel for it. I’ve been doing it too long.”
“But what I want to know from him, when it’s not a feel, when it, no, this is 98%. We don’t even need to talk about it. So we don’t need to talk about it. Tell me when it’s that, and I’ll just do it.”
Ron Yurko retweeted
Was lucky to research the relationship between linebackers covering ground in the run game and not biting on play action with @ericeager_ before he joined the Panthers. Most linebackers have to make a trade-off on flowing in the run game vs. staying back when the run fake is happening. Very cool to see the research is being extended on!
New stat(s), a request from @Foxworth24: run-pass reaction for linebackers!
Here is how quickly linebackers start to move downhill on run plays and how much vertical ground they cover in the first 1.0 seconds.
(Data: NFL Next Gen Stats)
Your votes this week count for more and are getting more valuable as we cross mid-season! Jeremiath Smith is closing in on Chambliss for the lead hopefuls.heisman.com/ #HeismanHopefuls #HeismanFanVote #CMSAC
Ron Yurko retweeted
The Week 7 #HeismanHopefuls Fan Vote leaderboard is live! 🔥
Vote now: hopefuls.heisman.com/ballot
Ron Yurko retweeted
For the first time, your vote will be the first official ballot for the Heisman Trophy. 🗳️
Vote now: hopefuls.heisman.com/ballot
Great paper on this subject by @RyanBrill_ & @adiwyner arxiv.org/abs/2311.03490 #analyticshavesomehumility
YES!
This is a great take on the subject!
We talked about this on the show a few weeks back.
You can't just treat it like poker or blackjack, cos you don't necessarily get the time for the math to work in your favor.
What's that line, the market can remain irrational longer than you can remain solvent? Same idea.
You want the information; you don't want to go with the edge lean every single time just cos it says 51%.
Ron Yurko retweeted
This is a VERY sweet way to both show observed data and the "shrunken" (estimated talent) data
I don't think I've ever seen that before?
Let me tag a few interested folks @Stat_Ron @StatsbyLopez @SchuckersM
Here's a post where I talk about the process of writing a baseall research paper using Claude:
baseballwithr.wordpress.com/…
Ron Yurko retweeted
Steelers 26, Buccaneers 19. Oct. 18, 2026.
Tampa Bay (41.5%) reached 91.0% win probability, up 16–3 with 0:00 left in Q3.
Under our benchmark, a collapse this extreme occurs in 9.93% of games. For teams starting at 41.5%, 7.03% of losses are this extreme.
WP: @ESPN
ALT A line chart shows Tampa Bay Buccaneers win probability throughout the game against Pittsburgh Steelers on Oct. 18, 2026. Final score: Pittsburgh Steelers 26, Tampa Bay Buccaneers 19. The highlighted peak is 91.0%, up 16–3 with 0:00 remaining in Q3. The first reported win probability was 41.5%. Source: ESPN.
Ron Yurko retweeted
Thanks so much to Ben for sharing this data publicly for the 2027 #BigDataBowl!
Grinding the Mocks aggregates mock draft data to create consensus-based projections of where prospects are likely to be selected in the NFL Draft.
Depending on focus, may be a very helpful resource.
Announcing a public @GrindingMocks dataset for use in the 2027 #BigDataBowl! Include mock draft/big board sentiment in your models and...prosper???
grindingthemocks.substack.co…
Ron Yurko retweeted
Brendan McCullough Becomes Carnegie Mellon’s All-Time Reception Leader in Program’s 600th Victory #TartanProud #d3fb
athletics.cmu.edu/x/oa4gc
