Adi Wyner
@adiwyner
Professor of Statistics and Data Science. Co-Faculty director of Wharton Sports Analytics and Business Initiative. co-host of Wharton Moneyball.
1 986Followers
305Following
927Posts total
20.7KViews on collected posts
Neueste Beiträge
Wir suchen neue Beiträge…
@adiwyner @tejfbanalytics @RyanPaganetti The data doesn’t explain their intentions it only explains whether it was effective if indeed they did it intentionally
390 views · 2 likes
· 0 reposts · 1 replies
Open on X →
11/11 Credit to @AlexPrejean, whose analysis of the Patriots’ kicking gap helped frame this thread. We add attempt-level expectations, home-road comparisons, clustered tests and FDR control: https://t.co/8SbxNzFroi
561 views · 4 likes
· 0 reposts · 3 replies
Open on X →
9/11 Blocking: NE opponents had 3.86% of attempts blocked versus 2.15% elsewhere (unadjusted p=.010), accounting for roughly one-third of the missing makes. Remove every block and opponents still finish 4.07 points below expectation.
495 views · 2 likes
· 0 reposts · 1 replies
Open on X →
10/11 Verdict: the data show an unusual NE opponent-FG effect—not compelling evidence of a home-only dirty trick. Blocking, weather, kicker mix and special-teams factors? Paganetti’s finding is fun to ponder, but could have happened without the video.
551 views · 3 likes
· 0 reposts · 2 replies
Open on X →
The home field effect was larger: −6.46 points at home versus −3.41 on the road. Statistically, no: the home-road difference has p=.38. The data cannot isolate a special Gillette-only mechanism with strong evidence. 7/11
521 views · 3 likes
· 0 reposts · 1 replies
Open on X →
8/11 Yet, across all venues, NE opponents went 398-for-518: 76.8% versus 81.8% expected—a −4.94-point gap and 25.6 fewer makes. Clustered p=.0012; 32-team FDR q=.0386. NE was the only FDR-significant negative effect.
497 views · 2 likes
· 0 reposts · 1 replies
Open on X →
5/11 At home, NE opponents went 196-for-260: 75.4% versus 81.8% expected. The −6.46-point residual was the largest home-opponent suppression among 32 teams. Clustered p=.0016—but after two-sided BH-FDR control, q=.0526: borderline.
552 views · 2 likes
· 0 reposts · 1 replies
Open on X →
On the road, home kickers facing NE went 202-for-258: 78.3% versus 81.7% expected. The −3.41-point residual was also the league’s lowest, but it was not significant: clustered p=.139 and FDR q=.701. 6/11
535 views · 2 likes
· 0 reposts · 1 replies
Open on X →
2/11 Paganetti reports that Gillette’s video board showed a reverse view with offset yellow uprights during opponent kicks. That's a testable prediction: NE’s suppression should be distinctly stronger at home than on the road after adjusting for factors that effect kick.
782 views · 4 likes
· 0 reposts · 1 replies
Open on X →
4/11 NE’s kickers were excellent: 533-for-618, or 86.2%, versus 84.5% expected (Vinatieri and Gostkowski !! ). But the overall +1.77-point edge was not huge or significant (clustered p=.22).
586 views · 3 likes
· 0 reposts · 1 replies
Open on X →
3/11 I modeled all 18,461 regular-season FG attempts from 2001–19. Each kick’s expected probability accounts for distance, indoor/outdoor status, their interaction, and season. I then tested team effects with standard errors clustered by game.
664 views · 3 likes
· 0 reposts · 1 replies
Open on X →
Did the Patriots use a dirty home-field trick on opposing kickers? @RyanPaganetti’s video-board finding says yes. The data says: not so fast. The home-road difference isn’t significant, blocks explain part, and home-only evidence is borderline 🧵 1/11 https://t.co/M8w6i2uLK4
14.5K views · 61 likes
· 12 reposts · 3 replies
Open on X →
Im Vergleich zu Konten gleicher Größe
12 Beiträge aus den letzten 90 Tagen, verglichen mit der Größenklasse under 10K Follower. liegt bei Reichweite und Interaktion unter vergleichbaren Konten.
Medianaufrufe552dieses Konto4 175Median für under 10K
Reichweite, %27.77%dieses Konto250.82%Median für under 10K
Interaktion, %0.62%dieses Konto1.41%Median für under 10K
| Kennzahl | Dieses Konto | Median für under 10K | Verhältnis |
|---|---|---|---|
| Medianaufrufe pro Beitrag | 552 | 4 175 | 0.13× |
| Reichweite (Aufrufe ÷ Follower) | 27.77% | 2.5× audience | 0.11× |
| Interaktionsrate | 0.62% | 1.41% | 0.44× |
Weitere Konten dieser Größe → Mit einem anderen Konto vergleichen → Wie diese Vergleichswerte entstehen →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
14.5K3 Sep
664
586
782
535
552
497
521
551
495
561
390
Last 12 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.54%3 Sep
0.60%
0.68%
0.64%
0.56%
0.54%
0.60%
0.77%
0.91%
0.61%
1.25%
0.77%
Reactions — likes, reposts, replies and quotes — divided by views. Median for under 10K accounts is 1.41%.
What the audience does
Likes63.2%91 in total
Reposts8.3%12 in total
Replies11.8%17 in total
Quotes2.1%3 in total
Bookmarks14.6%21 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
Ähnliche Konten
G2 jtay ✓
@JTayy_
4K followers
· 30.4K posts
sakhiyee
@sakhyiee
4K followers
· 10.3K posts
B1. ✓
@chicagoBRIZ
4K followers
· 197.1K posts
Grace Bonny ⬡
@godspower_123
4K followers
· 7K posts
Solid.xyz ✓
@SolidYield
4K followers
· 440 posts
AltOnChain ✓
@AltOnChain
4K followers
· 2.8K posts
Michigan Tech MBB
@mtumbb
4K followers
· 4.6K posts
Chris Ogilvie
@OgilvieCJ
4K followers
· 36K posts