Walker Harrison
@WalkWearsCrocs
Followers
1,487
Following
2,008
Media
361
Statuses
1,826
Now: quantitative analysis, @Yankees // Then: MS in statistics @Duke , content for @IBM , @ThePracticalDev // Always: blogging about data, missing lefty layups
New York City
Joined August 2013
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Russian
• 1085454 Tweets
Putin
• 574415 Tweets
Rich Homie Quan
• 404515 Tweets
Venezuela
• 326678 Tweets
Cristiano Ronaldo
• 225277 Tweets
Messi
• 147076 Tweets
San Marino
• 125253 Tweets
Chiefs
• 118162 Tweets
Hunter Biden
• 109899 Tweets
#السعوديه_اندونيسيا
• 90344 Tweets
Thug
• 83182 Tweets
Ravens
• 74983 Tweets
World Cup
• 70248 Tweets
#GHEstreno
• 59491 Tweets
Liechtenstein
• 49099 Tweets
From Zero
• 47749 Tweets
Linkin Park
• 44468 Tweets
كاس العالم
• 41742 Tweets
El Alto
• 38191 Tweets
Bolivia
• 35566 Tweets
مانشيني
• 34743 Tweets
Mahomes
• 33380 Tweets
سالم
• 31264 Tweets
Chester
• 21721 Tweets
المنتخب السعودي
• 21644 Tweets
店頭ビジュアル
• 21006 Tweets
Di María
• 20478 Tweets
Violeta
• 19507 Tweets
Emily Armstrong
• 16593 Tweets
Milk Marie
• 13526 Tweets
if you "pull the slack" out of a win probability chart, the length of the resulting line is a measure of the game's excitement. Top 4 games from Week 10:
36
108
2K
This flowchart shows my results from simulating NFL playoff overtime, a hot topic since the Super Bowl.
My numbers indicate that there is effectively no advantage between choosing to kick or receive, which aligns with research done by
@bburkeESPN
and
@StatsbyLopez
.
Details⬇️
79
178
1K
Much has been made about the divergence between
@FiveThirtyEight
's
#NBA
predictions this year and the corresponding Vegas lines.
For every '21-'22 regular season game, I tested how RAPTOR spreads and win probabilities compared to sportsbooks'.
full post:
5
22
228
if the NFL got rid of overtime and let games end as ties, there’d be way more variety in final record/win% and they wouldn’t need to determine the playoff pool with all those unsatisfying tiebreakers
full post:
15
24
167
If you could have bet against
@FiveThirtyEight
this year, you would have seen 20% - 30% ROI
#rstats
6
28
145
I made a neat lil GIF showing how kernel regression draws a smooth line
full story + code ->
1
48
115
I made this map of NYC that scores every block based on how much stuff (bars, restaurants, grocery stores, laundromats, etc.) is on it
full blogpost here:
made with
#rstats
, as always
motivation/implementation/caveats in thread below ⬇️
4
6
91
I'm also attaching this version of the chart with a legend, which is slightly less twitter-friendly. The methodology is:
-Receiving team kicks XP on first possession
-Kicking team goes for two down 7
-Defensive scores not shown (happens about 1% of time)
5
4
67
if you're curious how I made this animation, the code and explanation can be found on my blog:
1
3
57
the irony here is that our distaste for ties doesn’t actually eliminate them — it just shifts them from the game level and into the standings, where they’re much more consequential
3
6
52
three (3) whole friends asked me to pull the slack out of this one so I assume thousands of others feel the same…
second most “exciting” playoff game since 2011! only behind Ravens 38, Broncos 35 in 2012 (Jacoby Jones Hail Mary)
Jan 23, 2022
Final (OT): KC 42, BUF 36
Excitement Index: 7.413
Percentiles (Postseason games):
▪100.0% in 2021 (1 of 10)
▪99.1% since 2011 (2 of 122)
▪100.0% since 2011 for KC (1 of 15)
▪100.0% since 2011 for BUF (1 of 7)
#BillsMafia
#ChiefsKingdom
#BUFvsKC
2
7
53
1
5
46
But I do think that finding such a fundamental (and seemingly easily addressed?) flaw in their predictions makes it harder to buy into their
#NBA
forecasting methodology as a whole.
4
0
42
If your opponent scores 7 on the opening drive, you want to go for 2 because it's a 50% win proposition compared to 40% extending to sudden death where they have the ball first.
3
2
43
While I'm confident in my approach above, I'm sure there are legitimate critiques of the choices I made so I welcome any and all feedback. You can see the code here:
8
0
40
Some exciting personal news: will be working for the
@Yankees
this summer on their quantitative analysis team!
5
1
39
The assumptions shouldn't be too controversial. If you score on first drive, you don't want to go for 2 because then more than half of your opponents' responding TD drives will end up in losses (the 50% of the time you fail your 2-pt, and the ~10% they match and then win later).
5
1
39
me to my students: you need to write informative commit messages
me in real life:
4
6
32
the majority of value I get from making scripts public is that it forces me to write better code, even if no one ever sees it
Random thought: does anyone ever actually look at the code for
@ben_bot_baldwin
?
How valuable is open source if no one cares to look through the source code?
Or maybe the value is in being able to look if one wanted. idk
25
1
60
1
1
30
That doesn't mean the predictions are useless though. Once you remove the home team bias, a linear blend of sportsbook lines and RAPTOR lines actually slightly outperforms Vegas, at least for moneylines.
1
1
27
the key is to take each line segment and gradually lay it flat while also reattaching the endpoints in every frame
0
1
26
@NFLPinnacleBeat
@bburkeESPN
@StatsbyLopez
I agree with that. There are a lot of ways that accounting for team quality will budge these distributions one way or another.
But I find it unlikely that you could shift the probabilities enough to say that there’s a default, obvious choice.
3
0
26
If your defense is gassed then you might consider that first possession more score-heavy than reflected above; if you have one last trick play ready for your 2-pt conversion, maybe you don't mind having to go for 2; and so on
4
0
25
@DSP4150
@benbbaldwin
ha agreed...was hoping I'd get something a little more aligned with the experience. Ties get a big boost from the added overtime plays (and I don't normalize for game length)
3
0
24
The upshot is that both RAPTOR spreads and especially its win probabilities are biased toward the home team, and in particular for home underdogs. Home teams that RAPTOR thinks should win half the time are only winning about 35% of those games
#rstats
2
4
23
@DWAZ73
Sanchez played pretty well in that game. Cold road game against Pitt D and he turned 24-0 halftime blowout into 24-19 game. Needed the ball one more time but D couldn't get stop. Go kick rocks, Jason ✌🏻
2
1
19
The logic between nodes should also be pretty straightforward. Kicking teams down a touchdown will never attempt a field goal, XP are some times unnecessary, sometimes the difference between outcomes, etc.
1
0
22
these
#rstats
stickers were supposed to be for the undergrad students but IMO I deserve them more
(thanks
@minebocek
)
2
2
22
I think using the multi-line cursor is the coding equivalent of when the scissors start to glide through the wrapping paper 😩
#rstats
1
4
21
If OT was eliminated, there’d also be slightly higher correlation between point differential and win% because all those noisy one-score OT games would be “regressed” into half-wins
1
0
22
Simulating was done by stitching together drives from resampled plays in recent seasons using
@nflfastR
data. Constraints were added to make sure teams didn't do anything insane, e.g. kick a field goal down 7, and fourth down decisions were fed into
@benbbaldwin
's model.
4
0
21
my stat 101 students respect me so much that they photoshopped me as a dutch schoolgirl for their final presentations
1
0
20
as a cherry on top, getting rid of OT would save a handful of players from serious injuries every year
1
0
19
The key then is to figure out the transition probabilities between nodes. How do we determine the frequency at which teams will score 0, or 3, or 6-8 points? We can use a combination of empirical results and simulation.
1
0
18
The results illustrate how close to being 50-50 the choice to kick or receive is under reasonable assumptions, which gives coaches the right to have a preference if they feel confident that their particular situation skews the underlying distributions.
1
0
18
I miss New York
*sees on Instagram that today is SantaCon*
I don't miss New York
3
0
18
The victim in this crash was a friend of mine, Jim Pagels. We were supposed to chat soon to compare our cities' respective bikeshare programs.
Jim was a talented writer whose thoughtful, stats-oriented posts inspired me to try my own hand at blogging.
1/x
1
0
18
September: *starts data blog*
November: data fails everybody
December: ????
0
5
17
theme_bw looks way better than theme_gray and should be the ggplot default. I will die on this hill.
#rstats
0
1
16
Plugging in a blend of these simulations and empirical distributions yields the percentages you see above. Chaining together the probabilities and gathering them at the end-of-game nodes gives nearly a 50% win probability for each team.
1
0
16
For some nodes, simulation isn't necessary -- we have years of "old school" OT games to determine how often each team wins in a sudden death scenario. Other situations have little data (down 3 after the first possession), or none (down a touchdown after the first possession).
3
0
16
Back on twitter. need to either find a way to clear my ancient tweets or just push em down with empty filler content
#resurrectionproblems
1
0
15
me telling my mom I can't come to thanksgiving because I have to build a twitter bot for my internet friends
0
1
15
that’s odd since it’s my personal blog...can’t wait to meet them though!
2
0
15
new week, same idea -- straightening out the win prob graphs to find the most "exciting" (uncertain, volatile, etc.) games
plz don't yell at me if you didn't actually enjoy these games
1
0
15
happy valentine’s day or whatever. here’s a cool GIF I made
2
1
15
Recommender systems ranked:
1. YouTube’s sidebar
2. Spotify’s Discover Weekly
.
.
.
991. My dad sending me videos of tractors he likes
992. Whatever tf Netflix thinks it’s doing
2
0
14
revealed to me today: my colleagues have been playing bingo using my stupid catchphrases for the last few months
3
1
13
Wrote a summary of the newly released
#StackSurvey17
. Lots of notable trends to keep track of:
0
3
12
inside look at the reminders set by a recovering night owl
2
0
13
Excited to announce that I'll be presenting my research into NFL playoff overtime on May 16th at the 2024 New York R Conference (
@rstatsai
)
🎟️Tix & Info:
#rstatsnyc
|
#rstats
This flowchart shows my results from simulating NFL playoff overtime, a hot topic since the Super Bowl.
My numbers indicate that there is effectively no advantage between choosing to kick or receive, which aligns with research done by
@bburkeESPN
and
@StatsbyLopez
.
Details⬇️
79
178
1K
2
0
14
Nor does it really explain why they've been high on the Celtics and low on the Warriors this season.
2
0
12
I refuse to believe I live in a world where every ten minutes someone reports that ChatGPT painted a masterpiece or wrote an original symphony but identifying a half dozen vaguely distorted digits still counts as inimitable human intelligence
1
1
11
brought a mattress to school so I can start napping in the grad study room if anyone's wondering how I'm doing
2
0
12
@Tucker_TnL
I'm a little skeptical that Vikings still had 30% win prob on Minneapolis Miracle (no timeouts, :10 remaining). ESPN has it much closer to 0%
2
0
12
birthdays in 2021 are mostly just a reminder of how many institutions own your personal data
1
0
12
I don't actually know the difference between any of the `apply` functions in R, I just try them all until I get what I want
1
0
12
None of this is meant to denigrate RAPTOR, which was designed to evaluate individual players and not to predict games, or
@FiveThirtyEight
in general -- I've long admired a lot of the work that they put out.
1
0
12
@jacmoe
@ThePracticalDev
(Jeff) Atwood's Law: anything that can be written in JavaScript, eventually will be
#StackSurvey17
0
1
12
Jill Biden: has doctorate, rudely addressed as “kiddo”
Me: only has master’s, called “Dr.”, invited to speak at sketchy Russian AI conference
1
0
11
if The Rookie were set in the present, Dennis Quaid would need pinpoint control and a wicked slider to get away with throwing just 96
1
0
11
If your opponent scores 7 on the opening drive, you want to go for 2 because it's a 50% win proposition compared to 40% extending to sudden death where they have the ball first.
3
2
43
1
0
11
useful little
#rstats
tip
if you want to create a new ID/index of clusters of data (i.e. consecutive rows with same value in a column), you can simply take a cumulative sum of instances when the lagged value isn't the same:
0
2
10
always awkward when I have to explain to Safari how this is a casual, now-and-then thing and that it can't be my default browser...
1
0
11
@tejfbanalytics
I added that logic “just in case”…and then it happens the first week I test it
0
0
11
at a startup, every day brings new challenges. today I solved the Great Mysterious Cold Brew Leak 2017
1
0
10
Last night my friends and I attempted to take part in a timeless American ritual: watching a bunch of NCAA Tournament games at a sports bar.
Unfortunately we stumbled into one of the least competitive 2-hour stretches of basketball in recent March Madness Round 1 history. ⬇️
3
0
11
me compulsively refreshing the very thing that nearly gave me a heart attack 2 months ago
0
0
11
I even went to this block last week to tell its residents that they lived on the most convenient block in the city.
they did not care lol
1
0
11
Checked my assumptions & tried to not talk baseball w/ tall uber driver from DR. Turns out he pitched in MLB for 4 yrs.
We talked baseball.
1
0
10
flight attendant: Are you willing to sit in the emergency exit seat?
me: Sure haha just gotta kick open the hatch and parachute out right?
her: Seriously.
me: 😔 Yes ma'am.
1
0
11
tfw you’re about to get your little sister hooked on ggplot
0
0
11
do i need to wear goggles to put this ikea dresser together? no.
but do i want to?
yes.
1
0
9
my life goal is to write an entire book that's also a palindrome. I have the middle set, now I just need to work out in both directions
RACECAR
1
0
10
:') always nice to be reminded why I got back into the content game
1
0
10
@minebocek
@dgkeyes
pivot_wider was a complete betrayal of the hardcore fans who spent years memorizing how gather/spread worked
1
0
10
one of the unexpected joys of blogging was paying illustrators for original work.
here are a few of my favorites. DM me if you'd like any of their contact info
0
0
10
today's gripe: scoville units really need a log transformation
1
0
9
me wondering whether I can treat each of the three babies crying behind me on this plane as independent events as I calculate the probability I get any sleep whatsoever
0
0
10
If anyone's wondering what shitty spreadsheet version control looks like...
1
0
10
My twitter bio says “always blogging about data,” but in truth I haven’t posted new content in nearly 4 years.
So I’m getting back into it. Semi-regular content going up on . First post, exploring Google's autocorrect functionality, is live!
#rstats
2
4
10
this is a horrible perspective and I hope no one takes it seriously
1) Can I be one of the best in the world at this?
2) Do I really want to be?
If the answers are no, make it a hobby or quit.
33
54
606
1
1
9
me: shut up about your wordles
also me: here’s how quickly
#wordle
has grown this month based on how many people tweet out their little squares
3
0
9
good feature! my friends and I have been competing to find the most obscure answers everyday.
I do wonder if trying to maximize -log(x) would be better than minimizing sum(x) though. my instinct is that a 19% answer and a 1% answer is better than two 10% answers.
🚨 NEW: RARITY SCORE 🚨
After completing your game, you'll be able to see your rarity score on the summary page. The lower, the better! Shout out to
@DSzymborski
@FoolishBB
@JessicaDBrand
for inspiration
98
41
2K
1
0
9
Trying to find uphill walking routes shows power of relaxed constraints.
Under strict uphill requirements, I only get a few blocks before hitting local max. But under loosened constraints (elevation can decrease by 1 ft btwn blocks), I make it 50 blocks and finish at global max
1
0
8
pro tip: if you ever don't want to type out a gnarly LaTeX formula, find it on Wikipedia, look at the page's source code, and copy-paste 🤓
1
0
9
@jessleenyc
@ThePracticalDev
#stacksurvey17
would be interesting 2 normalize results to industry standards. Where are devs most undervalued?
1
0
9
when I first got into statistics and forecasting, I read Black Swan and loved it. I also read books like Signal & Noise and Nudge, and loved them too. Turns out the author of the former now spends his days childishly trolling the authors of the latter on Twitter.
what a waste
1
0
9
my first act of 2018 was drunkenly buying a burrito on the street but then leaving it in a cab so yeah gonna be a wack year
1
0
9
if your screenshot includes the other tabs that were open, those are the only thing I will look at
1
1
9
When your roommate misses a softball game, it's best practice to write up a petty fake recap
2
0
9