Sam Wirth
@SamWirthSports
Followers
877
Following
561
Media
58
Statuses
366
Georgetown '25 math, econ, history. Baseball projects @Mariners .
Seattle/Washington DC
Joined March 2022
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Kendrick
• 357343 Tweets
Super Bowl
• 297098 Tweets
Taylor
• 289522 Tweets
NYFW24 WITH GEMINI FOURTH
• 284963 Tweets
JISOO AT NYFW WITH TOMMY
• 158298 Tweets
Wayne
• 125409 Tweets
PONDPHUWIN NYFW2025
• 124194 Tweets
Browns
• 102551 Tweets
Cowboys
• 96424 Tweets
Watson
• 81132 Tweets
Lions
• 73571 Tweets
Raiders
• 57467 Tweets
重陽の節句
• 56518 Tweets
Chargers
• 54784 Tweets
Baker
• 41369 Tweets
ON CH3 iQIYI JACK JOKER
• 38300 Tweets
#JACKANDJOKER1stEP
• 37481 Tweets
Broncos
• 37102 Tweets
Cleveland
• 33279 Tweets
Tom Brady
• 29014 Tweets
Rams
• 27144 Tweets
#ZEEgotoNYFWwithCOS
• 26595 Tweets
ラヴィット
• 21114 Tweets
Alex Morgan
• 19317 Tweets
チャウヌ
• 19031 Tweets
もちづきさん
• 16890 Tweets
Bo Nix
• 14560 Tweets
#OnePride
• 13788 Tweets
ZEE x COS
• 11068 Tweets
ふっかさん
• 10396 Tweets
看護倫理学会
• 10296 Tweets
Pinned Tweet
Going to start publishing some papers I've written so far in college on Medium because I think they're cool. Check them out!!
3
0
13
I am excited to announce that I have accepted an offer to work as a baseball projects intern for the Seattle Mariners this summer. Having grown up in Seattle, this is a dream come true.
37
10
296
Check out my new article that explains the mathematical theory behind the XGBoost machine learning algorithm and provides a practical application with baseball data:
3
7
93
Interested in Vertical Approach Angle? Here is how to calculate VAA based on Statcast pitch-by-pitch data.
5
4
85
Publishing my Spring Training Shiny app! It has Statcast data for all pitchers and batters from the spring.
8
7
82
Here is the link to my Triple-A shiny app! It has pitch movement data for all 2023 Triple-A games.
1
8
51
Just updated the Triple-A and FSL app for the last time. Thank you to everyone who used it!
5
9
48
Now that Spring Training is over, I wrote a brief article explaining which spring stats matter and which don't. I found that due to a lack of sample size, batting stats have little value in predicting in-season success. On the pitching side...
2
6
45
I am very excited to announce that I have finished the documentation for my MiLB run-expectancy project and have published the code to GitHub and write-up to Medium and RPubs. The links for these are posted below. (1/2)
2
8
30
Massive update to the site:
- 2022 data for comparison
- Visuals showing velocity, spin rate, IVB, and HB by appearance
- Faster load times
2
1
28
Does vertical approach angle (VAA) matter?
Yes – but also no. While flatter fastballs perform better than ones with steeper approach angles, one cannot simply create a flatter fastball with their existing mechanics and pitch locations.
1
4
22
Check it out!
🚨🚨 AAA Statcast App by
@SamWirthSports
is now at Prospects Live 🚨🚨
Day 2 of DATA WEEK™️ brings easy of access to AAA Statcast leaderboards and player views for both hitters & pitchers. FSL will also be included!
Link:
Intro:
1
15
33
1
2
21
Georgetown starting pitcher
@nolan_mccarthy
had a fantastic bounce back outing today. 6 IP, 2 R, 0BB, 11Ks. Of his 90 total pitches, 71 were strikes. That'll play.
1
0
20
Which baseball stats actually matter? Check out the piece I did for the
@RainiersLand
DIRT magazine explaining useful metrics for new fans and analysts alike.
1
2
18
Some new features I've added to the app over the last month or so:
- batter performance vs pitch types
- barrel% and sweet spot%
- launch angle density plot
- optimized code for faster load times
2
2
16
Updated!
@SamWirthSports
I'm here to bug you again ... any way to get the pitch results like SwStr% or GB%?
1
0
0
2
2
17
I am new to R but am doing as much as I can to get better. Here I tested the correlation between a Major League Baseball team’s pitching
@fangraphs
WAR and their overall ERA, SIERA, FIP, and xFIP numbers from 2019 to 2021.
#rstats
1
3
14
Super excited about this! Had a lot of fun working on the stuff and command models in 2023.
We are proud to introduce two new tabs to our SYNC Tile: 𝙋𝙞𝙩𝙘𝙝𝙞𝙣𝙜 𝙈𝙤𝙙𝙚𝙡𝙨 & 𝙃𝙞𝙩𝙩𝙞𝙣𝙜 𝙈𝙤𝙙𝙚𝙡𝙨
Here users can interact with models such as Stuff+, Command+, xRV+, Swing Decision+, and Estimated Bat Speed🔥
Learn More:
0
0
10
1
1
16
Disgusting combo
Félix Bautista, 102mph Fastball and 91mph Splitter, Overlay
5
31
518
0
0
16
The site now has full movement!
Also, note that the IVB and HB are only in the last 40 feet (not full movement).
0
0
0
2
1
14
Georgetown Baseball has a record of 26-17 with 10 games remaining. The Hoyas need to win 1 more game to finish above .500, something that hasn't been done since 1986.
0
0
12
A few people messaged me about slow load times on the AAA app so I just optimized my code and everything should be way faster now
1
2
11
I updated my spring training shiny app for the last time today. Thanks to everyone who used it!
1
1
9
Canzone destroyed AAA this season:
113 max ev
108 95th pct ev
34.8 sweet spot%
8.9 barrel%
44.2 hard hit%
12.8 BB%
149 wRC+
The full deal, as
@Ken_Rosenthal
said, is Paul Sewald to the Diamondbacks, with big leaguers Josh Rojas and Dominic Canzone as well as minor league middle infielder Ryan Bliss heading back to the Mariners.
120
310
3K
0
0
10
Rogers' slider is arguably the most fun pitch in baseball. Here it is with the sinker that was thrown the pitch before.
0
0
10
Historic season for the Hoyas!
With tonight's win, the Hoyas have set the program's single-season record in wins, surpassing the 1986 team (29)!
#HOYASAXA
8
20
224
0
0
8
Attached is a spreadsheet containing the wOBA, wRAA and wRC+ of every Triple-A hitter this season (min 25 PA), sorted by team:
0
2
9
I had a blast attending the
#SABRanalytics
conference this weekend. Thanks to
@MLB
,
@SportsInfo_SIS
, the research presenters, and the panelists who made it all possible.
1
0
7
The best in the business
0
0
9
Does Called Strike Plus Whiff Rate (CSW) matter in evaluating a pitcher? CSW is a stat found by adding a pitcher’s called strikes and whiffs, then dividing by their total pitches thrown. As demonstrated by the graphs attached, CSW shows weak correlation with (1/3)
#rstats
1
1
7
2022 Triple-A team wOBA leaderboard:
1. El Paso Chihuahuas - 0.361
2. Albuquerque Isotopes - 0.351
3. Oklahoma City Dodgers - 0.349
.
21. Tacoma Rainiers - 0.319
.
28. Syracuse Mets - .303
29. Gwinnett Stripers - .302
30. Buffalo Bisons - .300
1
1
8
Bloss yesterday against St. John's:
- 41.5 CSW% (!)
- 26.6 SwStr% (!!)
- 73.5 Whiff% (!!!)
- -7.0 run-value (!!!!)
#DraftJakeBloss
HOYA W on Alumni Day!!
Hoya Highlights:
- Jake Bloss 7 IP, 1 Hit, 12 K’s 🔥🔥
- Marco Castillo Pinch-Hit 💣💣
-
@l_ub25
@austin_k12
@EzeMike23
& Angelo D’Acunto Multi-hit Games
#HoyaSaxa
|
#Team153
0
1
15
0
1
8
I think he's ready...
Fernando Tatis Jr. will work out at Petco Park the next three days and travel to Arizona with the team after Wednesday’s game.
The plan is for him to not play another minor league game.
(He is still suspended, so he will not be on the field by the time gates open to the public.)
11
35
514
1
4
8
Thank you so much to everyone who helped me along the way. This would not be happening without all of you.
0
0
8
During the
@Mariners
broadcast today,
@TheDaveSimsShow
mentioned that Andrez Munoz has thrown his slider harder recently which has made the pitch more effective. As shown by the graphs, Munoz has been throwing his slider with more velocity and this corresponds to an (1/3)
#rstats
2
2
7
Good luck...
Evan Phillips, 96mph Two Seamer and 87mph Sweeper, Individual Pitches + Overlay
5
15
277
0
0
7
Georgetown students, I am happy to announce that
@GtownAnalytics
is hosting
@NikOza2
in ICC 106 at 8:30pm tomorrow. He graduated from Georgetown in 2016 and is currently the Director of Baseball Analytics at Big League Advance. He also interned with the Dodgers, Rays, and Suns.
2
0
6
Nolan McCarthy (
@nolan_mccarthy
) yesterday:
9IP, 5H, 0R, 0BB, 12K, 115 pitches
McCarthy this season:
3.94 ERA
-.088 FIP (unadjusted)
.304 wOBA
23.6 K%-BB%
COMPLETE GAME SHUTOUT!!! Nolan McCarthy gives the Hoyas their 25th win of the year!!!
#HOYASAXA
|
@d1baseball
1
18
125
0
0
6
I found that VAA is a byproduct of pitch release height and pitch location height, and thus is not a statistic that a pitcher can generate in the same way that one generates velocity or spin rate.
2
0
7
2022 Triple-A wOBA constants:
League wOBA = 0.325
wOBAScale = 1.16
wBB = 0.70
wHBP = 0.72
w1B = 0.84
w2B = 1.20
w3B = 1.47
wHR = 1.90
0
0
6
Updated statistics following Georgetown’s loss to No. 24 Maryland.
1
0
5
Last update to the site (I think...):
- xBACON and xwOBACON
- Barrels
- Max distance
0
0
6
Tacoma Rainiers wRAA leaderboard:
1. Jarred Kelenic - 9.99 (248 PA, 149 wRC+)
2. Sam Haggerty - 5.50 (179 PA, 142 wRC+)
3. Brian O'Keefe - 4.80 (254 PA, 133 wRC+)
4. Kyle Lewis - 4.59 (60 PA, 176 wRC+)
5. Mitch Haniger - 4.02 (33 PA, 209 wRC+)
1
0
5
0
1
4
Using the first slider to set up the second
0
0
6
Georgetown's Nolan McCarthy threw a complete game today, giving up 2 runs and striking out 4. This season, he has pitched incredibly well:
.282 wOBA (opposing hitters)
-.005 FIP (unadjusted)
18.8 K%-BB%
0
0
5
Just witnessed my first MiLB in-game fan proposal. Hopefully it is my last.
0
0
4
Not the result we wanted but putting up a serious fight against the
#11
team in the nation is something that this program should be proud of.
0
0
4
Jake Hyde is a name to learn. The Georgetown sophomore is slashing .440/.505/.833 in 20 games with an advanced power tool and great plate discipline. Hyde has hit for extra bases in 20% of hits at-bats while only striking out 9.5% of the time.
0
1
4
Updated statistics following Georgetown’s series against Xavier.
1
1
3
situationally significant moments, I hope to create a model that will predict which pitcher will perform the best in each possible scenario a team could face. (2/2)
0
0
3
Jhoan Duran with impressive stuff in his first outing of the spring.
0
0
5
Hearing rumors that Gonzaga is looking to join the Big East. Check out the article I wrote about potential conference expansion back in November.
1
1
4
Right now, my situational stats merely reflect the outcomes of past games. In the coming months, I look to learn how to incorporate machine learning into my model in order to predict the outcome of future games as well. Instead of just stating who has pitched well in (1/2)
1
0
4
A program record 9 home runs in a game. Looking very strong headed into Big East play this weekend.
1
0
3
Congratulations to
#Cardinals
slugger Albert Pujols on becoming the fourth member of the 700 Home Run Club!
#RStats
#ggplot2
0
2
3
Significant than one in the 9th inning, so the plate appearance in the 9th inning will be weighted more heavily. (2/2)
0
0
3
Keller's numbers from today. Also, it appears that "sweeper" got added as a pitch classification type.
This MIGHT be the revamped cutter Mitch Keller was talking about.
90 mph. 28" of vertical break.
13
8
289
2
0
4
Please note: This solely uses Triple-A batted ball data. The wOBA weights come from my Triple-A run-expectancy matrix.
2
0
4
Hop on the Jake Hyde bandwagon while you still can
0
0
3
For context, the “s” before a statistic refers to that statistic being situationally weighted. For these, each plate appearance is multiplied by a linear weight determined by its situational significance. For example, a plate appearance in the 1st inning is far less (1/2)
1
0
3
Updated statistics following Georgetown’s loss to No. 24 Maryland.
0
0
3
future performance of a pitcher. This is definitely something that I plan on looking into further.
Note: The data used are taken from the 59 qualified pitchers on both Fangraphs and Baseball Savant from the 2022 season. (3/3)
0
0
3
I’m going to start tweeting the recaps I write for the Rainiers
1
1
3
Definitely check this out. Really great stuff.
Turned wOBA table into Advanced Offensive Metrics table, adding Batting Average on Balls in Play (BABIP), Isolated Power (ISO), and "basic" runs created (RC):
@GtownBaseball
0
0
1
1
0
2
metrics that reflect current performance (fWAR, ERA, FIP), meaning that it should not be used to evaluate the performance of a player thus far in the season. However, due to its stronger correlation with xFIP, a predictive stat, CSW could potentially be used to project the (2/3)
1
0
3
approach with runners in scoring position at the Triple-A level. (2/2)
0
0
4
Cal Raleigh in June 2022:
.250/.337/.588 (.924 OPS)
14 XBH
.389 wOBA
163 wRC+
Mike Zunino in June 2017 ("Junino"):
.304/.371/.722 (1.092 OPS)
13 XBH
.443 wOBA
186 wRC+
@TheBGustafson
Gotta look this up, but off the top of my head, this has to be the best month by an M’s catcher since Zunino’s June from years back, right?
5
4
85
1
0
4
At its peak, it was getting 700+ hours of usage a month which was super cool to see. Should be rolling out a fun new app eventually, but that is very dependent on school workload
0
0
4
uptick in
@fangraphs
slider runs above average per 100 pitches (wSL/C). However, due to the incredibly small sample size (only 27 total appearances on the season), one cannot difinitively say that the increase in slider velocity is the reason for the increase in (2/3)
1
0
3
It is worth noting that the run expectancy with a runner on first and none out is higher than the run expectancy with a runner on second or third and none out. This could be a product of lower sample size (only a half season of data) or simply reflect the inferior hitting (1/2)
1
0
4
Application of the Hegelian State: Lassalle’s Focus on the German Proletariat and the Marxist Misconstrual
1
0
3
1
1
2
I wrote an article about Georgetown’s historic swimming and diving season.
0
0
2
0
2
2