Cardinals Simulation, September 25th

by Xeifrank on Sep 25, 2009 1:11 AM EDT

I used my simulator to simulate 50000 baseball games between the Cardinals and Rockies using what to the best of my knowledge will be close to friday's starting lineups. The simulator outputs a win probability for each team, along with the average total runs scored, a distribution matrix of all the final scores and how often they occurred, player stats, and an averaged box score tally of all games combined.

Today's Results... (Last simulation ran at 10:00PM)

Visitors	Home	Pitching Matchup	Favorite	Vegas Win Prob	Simulator Win Prob	AccuScore
STL	COL	C.Carpenter vs A.Cook	STL	58.85%	61.17%	60%

Skinny: Simulator Fun Facts... Both my simulator and the Vegas ML in good agreement on the Cardinals having around a 60% chance of winning this game. Vegas seems to like Aaron Cook, who hasn't pitched since August 21st, a little more than my simulator does. It will an interesting game as the Rockies could very well be the Cardinals first round opponent. The Rockies will want to show the Cardinals that they can hit against the Cardinals ace.

Top 20 Most Likely Scores
1	COL 4-3
2	COL 3-2
3	COL 5-4
4	STL 4-3
5	STL 3-2
6	STL 4-2
7	COL 2-1
8	COL 6-5
9	STL 5-4
10	STL 5-2
11	STL 5-3
12	STL 3-1
13	STL 6-2
14	STL 6-5
15	STL 6-3
16	STL 6-4
17	STL 4-1
18	COL 7-6
19	COL 4-2
20	COL 5-3

Game Pitching Results
Pitcher	WHIP	FIP
C.Carpenter	1.258	2.743
A.Cook	1.476	4.538

Note: Keep in mind I did this simulation before knowing the actual starting lineups, so some of the players I used may not be starting. Many of the "Players Most Likely To" stats depend on having the correct lineup. The lineups I used are listed below. Chances are you won't see a big difference in win probability from having the lineup order slightly off. Picking the correct starters is more important, especially if one of the better players is taking the day off. But you still won't see a big change in win probability if the starter and backup are interchangeable.

Simulation Lineups
1	R.Schumaker	C.Gonzalez
2	B.Ryan	S.Smith
3	A.Pujols	T.Helton
4	M.Holliday	T.Tulowitzki
5	R.Ludwick	B.Hawpe
6	R.Ankiel	Y.Torrealba
7	M.Derosa	I.Stewart
8	Y.Molina	C.Barmes
9	C.Carpenter	A.Cook

Tweet Comment 23 comments | 1 recs |

Do you like this story?

Comments

lol

colby doesn’t even get to start in a simulated game.

Interesting that the top 3 most-likely scores all have the rockies on top (in one-run games), but the cardinals win more games overall.

it's Clydesdales vs Goats. Actually sums up Cards vs. Cubs quite nicely. -all4tookie

by SleepyCA on Sep 25, 2009 2:09 AM EDT reply actions

I expect that Coors

would better benefit a hitter like Rasmus over Ankiel.

Guys like Bradley are exactly why we can't have a pumpkin patch anymore.

by liam on Sep 25, 2009 9:35 AM EDT up reply actions

I don’t follow the Cardinals much so the lineup may be slightly off but as long as I get the big guns in the lineup correctly, it shouldn’t have much effect. The home team almost always has the top two or three most likely scores just based on the rules of baseball (walks offs, not batting in the B9th with a lead etc…)

Thanks for the thoughtful comments.

vr, Xei

by Xeifrank on Sep 25, 2009 10:30 AM EDT up reply actions

Xei

Are you planning on running the sim prior to the playoffs getting started?

Future Redbirds - tracking Cardinal prospects for Cardinal Nation

by azruavatar on Sep 25, 2009 11:28 AM EDT reply actions

I will run the sim for all playoff games. If you guys would like I can create a FanPost for your playoff games like I do over at TBLA.
vr, Xei

by Xeifrank on Sep 25, 2009 12:37 PM EDT up reply actions

that would be awesome.

by WyoCardsFan on Sep 25, 2009 1:23 PM EDT up reply actions

Yes, that would be totally sweet.

Albert Pujols does not have "down" years. He has "~6 WAR" years.

by mattybobo on Sep 25, 2009 2:17 PM EDT up reply actions

and if you’d all like a longer list than “Top 20” for the most likely scores then let me know. I have to type them all in, so I don’t want to make it too long. 20-40 is reasonable. We had a Dodger game vs the Nats, which we won something like 15-2 (Hey, who hasn’t beaten the Nats 15-2 this year, right?) and it was something like the 128th most likely score. I can even output expected box scores, but until I have that automated, that’s a bit too much. Maybe if you end up playing the Dodgers in one of the series, I can output more data.
vr, Xei

by Xeifrank on Sep 25, 2009 5:14 PM EDT up reply actions

The most likely scores are kind of interesting, but whatever you are willing to do I'd be satisfied with.

Albert Pujols does not have "down" years. He has "~6 WAR" years.

by mattybobo on Sep 26, 2009 11:27 AM EDT up reply actions

I really like the sims

but I, for one, am really not that bothered about the “most likely scores” thing. I can’t see it being that interesting, particularly, other than for curiosity value. The simulations are cool though, and it’d be great if you could fanshot them in the post-season!

Felonius Monk - bitching to contact since 2008

by Felonius_Monk on Sep 28, 2009 5:04 AM EDT up reply actions

BCB'd

NO NEED TOO PLAY THE GAME NOW TEH SABERMAGICIANS HAVE SPOKEN DURRRRR

by Andyfantastic on Sep 25, 2009 12:09 PM EDT reply actions

R. Schumaker?

Ray? Rob? Randy?

V, b.

by LukeMP1186 on Sep 25, 2009 12:23 PM EDT reply actions

haha, fat fingered that’s all.
vr, Xei

by Xeifrank on Sep 25, 2009 12:28 PM EDT up reply actions

Yeah, should be "J" if anything....

Felonius Monk - bitching to contact since 2008

by Felonius_Monk on Sep 28, 2009 5:04 AM EDT up reply actions

I have no clue how this stuff works...

But I find it interesting that despite the Cardinals having a better than 60% chance of winning, the top 3 most likely score outcomes are in favor of Colorado.

V, b.

by LukeMP1186 on Sep 25, 2009 12:25 PM EDT reply actions

I will try another quick explanation on why this is. First let me lay the ground work with a couple of facts.

1. Most games are decided by one run.
2. Cardinals should win around 61% of the games in the sim.
3. Home team bats last and if they score a walk off run that puts them up by one run, the game ends, they don’t keep batting to run up the score.
4. When the visiting team takes a one run lead in the 9th inning or later, they keep batting and can increase their lead. Home team can’t.

What we end up with is even though a visiting team may be favored, as the Cardinals are 61% faves in this case. The most likely final game score will almost always be the home team winning by one run. Something like 3-2,4-3,2-1,5-4 depending on the over/under for that game. The final scores that the visiting team wins by is more spread out over many different run differentials (1, 2, 3 being the most likely), where the final scores that have the home team winning are more concentrated around 1 run differentials.

Simple example.
Final scores…
COL 4-3
STL 4-2
STL 4-3
STL 3-2
COL 4-3
-——————
STL wins 3 out of the 5 games, or 60% of the games, yet the most common score is COL 4-3. This is an extremely simplified example of what is going on. Hope this helps.

vr, Xei

by Xeifrank on Sep 25, 2009 12:35 PM EDT up reply actions

is that trend visible in real game data as well?

seems like it should be if the explanation you gave is the main effect.

Colby Rasmus nickname generator (as told by TLR)

by brackenthebox on Sep 25, 2009 1:00 PM EDT up reply actions

Yes, I’m sure it is. But I will let you pick through the empirical data and get back to us with the results. :)
vr, Xei

by Xeifrank on Sep 25, 2009 1:04 PM EDT up reply actions

as you said, it's there in real games too (at least this year)

http://www.vivaelbirdos.com/2009/9/25/1055084/the-clutchiness-of-the-home-team

Colby Rasmus nickname generator (as told by TLR)

by brackenthebox on Sep 25, 2009 5:53 PM EDT up reply actions 1 recs

Excellent work

Thanks for doing that. You came through in the clutch! :)
vr, Xei

by Xeifrank on Sep 25, 2009 6:10 PM EDT up reply actions

That explanation was very helpful I thought.

Basically the specific scoring outcomes are not all that important if all you’re interested in is who is more likely to win, period. They are just the various possibilities, or data points, or whatever, the sum of which you use to come up with a final probability of the original either/or (Cards win/Rockies win) scenario.

Albert Pujols does not have "down" years. He has "~6 WAR" years.

by mattybobo on Sep 25, 2009 2:20 PM EDT up reply actions