NCAA Game Simulator Prefers UM over Sparty
So this here NCAA Game Simulator is fun, even though what goes into the algorithm is unclear. My results:
FINAL SCORE -- Michigan (2016): 28 - Michigan State (2016): 21
Matchup Simulation Stats
- This game has been simulated 152 times.
- Michigan (2016) has won 134 times (88.2%), won by > 20 pts 70 times (46.1%)
- Michigan State (2016) has won 18 times (11.8%), won by > 20 pts 0 times (0.0%)
- The average score is Michigan (2016): 27.2 - Michigan State (2016): 9.6, decided by < 5 pts 19 times (12.5%)
- The game has gone into overtime 2 times (1.3%)
[Note: when it says "2016" it means "2015-16"]
[Also note: I ran the sims for this matchup with teams from 13-14 (84% to Sparty) and 14-15 (75% to Sparty), so the sim favors us to roughly the same degree it favored them the last two times...and we all know how that ended up.]
October 12th, 2015 at 7:05 PM ^
This is relevant to my interests.
Along the same lines, ESPN's FPI (I know, I know) gives us a 73.1% chance of winning.
October 12th, 2015 at 7:05 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 7:08 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 7:07 PM ^
October 12th, 2015 at 7:11 PM ^
October 12th, 2015 at 7:12 PM ^
2016? Who are the starting QBs for both teams?
October 12th, 2015 at 7:12 PM ^
No reason to really play the game. The data is in and its clear.
October 12th, 2015 at 7:14 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 8:28 PM ^
If they are doing it correctly they are taking into account each play from scrimmage. So starting from the opening kickoff, they simulate each play of the game based off percentages. Since they are based off percentages the outcome of each play and how many plays in every game will be completely diffrent. The more data they have the more accurate the simulation.
October 12th, 2015 at 7:16 PM ^
Over at Massey Ratings, we've got an estimated 69% chance with a projected mean score of 20-13 over their simulations. It's just funny to think that, at the beginning of the season - and granted, with no data from this season in the algorithm yet - that projection for that same game? It was markedly less rosy.
October 12th, 2015 at 7:22 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 7:24 PM ^
HARBAUGH
October 12th, 2015 at 7:28 PM ^
to set up the pass, we're pretty damn good at stopping the run, they will be in 3rd and long a lot, not a good recipe.
No way we give up 3 TD's when we've only given up 2 in the last 5 games, we could easily have had 5 shutouts in a row, that's not a fluke.
We're going to win this game by multiple scores with MSU not scoring very much, maybe 13 points or something, if that.
Meanwhile MSU's defense has conceded quite a lot of points and yards and for the 1st time in a while we will be the tougher team and smack them in the mouth.
October 12th, 2015 at 7:47 PM ^
MSU had 132 yards rushing, but most of their yardage came on big plays. Take away each RB's long and it was slim pickings against RUTGERS
CAR | YDS | AVG | TD | LONG | |
---|---|---|---|---|---|
Gerald Holmes | 5 | 37 | 7.4 | 0 | 30 |
Madre London | 15 | 27 | 1.8 | 0 | 8 |
EDIT: Chart didn't quite work...blerg.
October 12th, 2015 at 7:55 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 8:22 PM ^
The "remove the big plays" caveat is applied far too often on this here board, and for no good reason. "If I remove the data points that don't support my argument, my argument becomes valid." I don't believe this is a prudent way to analyze data. The plays actually happened, and are therefore equally important to the data set.
Holmes didn't recieve good blocking on his carries of 1 and 3 yards, so i'm going to remove them from my analysis. Thus he actually averaged 11 yards per carry last week, and should promptly be considered a Heisman candidate.
October 12th, 2015 at 7:29 PM ^
I think those projections are out of date. If you look at the simulator for the OSU game they win 80% of the time.
October 12th, 2015 at 7:33 PM ^
I'm just going to leave the result when I simulated it here and make no additional comment...
FINAL SCORE -- Michigan (2016): 38 - Michigan State (2016): 0
October 12th, 2015 at 7:36 PM ^
...good sir.
October 12th, 2015 at 7:38 PM ^
So do we still play the game or no?
October 12th, 2015 at 7:38 PM ^
Ran it once.
FINAL SCORE -- Michigan (2016): 17 - Michigan State (2016): 10
Rudock 17/30 2 TD 3 INT
Cook 11/27 1TD 1INT
LJ Scott 15 carries 73 yds
De'Veon Smith 20 carries 101 yds
Kerridge 2 carries 39 yds
October 12th, 2015 at 7:41 PM ^
They are everywhere right now.
October 12th, 2015 at 8:25 PM ^
Rudock throws three picks and we still walk away with a win, sign me up!
October 12th, 2015 at 7:40 PM ^
"FINAL SCORE -- Utah (2016): 14 - Michigan (2016): 34
Matchup Simulation Stats
- This game has been simulated 30 times.
- Utah (2016) has won 6 times (20.0%), won by > 20 pts 0 times (0.0%)
- Michigan (2016) has won 24 times (80.0%), won by > 20 pts 12 times (40.0%)
- The average score is Utah (2016): 7.6 - Michigan (2016): 22.9, decided by < 5 pts 5 times (16.7%)
- The game has gone into overtime 1 times (3.3%)"
Oh well, that was fun while it lasted. . .
October 12th, 2015 at 7:45 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 8:07 PM ^
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 8:09 PM ^
FINAL SCORE -- Michigan (2016): 6 - Michigan (2016): 38
Matchup Simulation Stats
- This game has been simulated 18 times.
- Michigan (2016) has won 8 times (44.4%), won by > 20 pts 2 times (11.1%)
- Michigan (2016) has won 10 times (55.6%), won by > 20 pts 3 times (16.7%)
- The average score is Michigan (2016): 18.8 - Michigan (2016): 20.2, decided by < 5 pts 3 times (16.7%)
- The game has gone into overtime 2 times (11.1%)
Hmmm. .
October 12th, 2015 at 7:40 PM ^
October 12th, 2015 at 7:54 PM ^
FINAL SCORE -- Michigan (2016): 35 - Michigan State (2016): 0
I'll take it.
October 12th, 2015 at 8:02 PM ^
Fuck it.
Michigan wins, 42-0
October 12th, 2015 at 8:04 PM ^
October 12th, 2015 at 8:08 PM ^
I just simulated us vs Alabama A&M and we supposedly had to punt 10 times!
http://www.ncaagamesim.com/FB_GameSimulator.asp?HomeTeam=Alabama+A%26M&…
Of course, they punted 15 times. Was this game simulated in a blizzard?
October 12th, 2015 at 8:15 PM ^
FINAL SCORE -- Michigan (2016): 6 - Michigan (2016): 23
Matchup Simulation Stats
- This game has been simulated 19 times.
- Michigan (2016) has won 8 times (42.1%), won by > 20 pts 2 times (10.5%)
- Michigan (2016) has won 11 times (57.9%), won by > 20 pts 3 times (15.8%)
- The average score is Michigan (2016): 18.1 - Michigan (2016): 20.3, decided by < 5 pts 3 times (15.8%)
- The game has gone into overtime 2 times (10.5%)
Michigan with the win!
October 12th, 2015 at 8:17 PM ^
What would this simulator have predicted for the NW score?
October 12th, 2015 at 8:20 PM ^
FINAL SCORE -- Michigan (2016): 24 - Northwestern (2016): 0
Matchup Simulation Stats
- This game has been simulated 323 times.
- Michigan (2016) has won 229 times (70.9%), won by > 20 pts 54 times (16.7%)
- Northwestern (2016) has won 94 times (29.1%), won by > 20 pts 6 times (1.9%)
- The average score is Michigan (2016): 17.5 - Northwestern (2016): 11.0, decided by < 5 pts 99 times (30.7%)
- The game has gone into overtime 14 times (4.3%)
October 12th, 2015 at 8:32 PM ^
October 12th, 2015 at 8:41 PM ^
FINAL SCORE -- Michigan (2016): 51 - Michigan State (2016): 0
Matchup Simulation Stats
- This game has been simulated 457 times.
- Michigan (2016) has won 414 times (90.6%), won by > 20 pts 224 times (49.0%)
- Michigan State (2016) has won 43 times (9.4%), won by > 20 pts 1 times (0.2%)
- The average score is Michigan (2016): 28.5 - Michigan State (2016): 10.1, decided by < 5 pts 54 times (11.8%)
- The game has gone into overtime 7 times (1.5%)
Please, stop, I can only get so erect.
October 12th, 2015 at 8:59 PM ^
Matchup Simulation Stats
This game has been simulated 275 times.
Michigan (2016) has won 157 times (57.1%), won by > 20 pts 44 times (16.0%)
Ohio State (2016) has won 118 times (42.9%), won by > 20 pts 21 times (7.6%)
The average score is Michigan (2016): 24.0 - Ohio State (2016): 21.1, decided by 5 pts 59 times (21.5%)
The game has gone into overtime 14 times (5.1%)
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 9:00 PM ^
FINAL SCORE -- Michigan (2016): 20 - Ohio State (2016): 10
Sent from MGoBlog HD for iPhone & iPad
October 12th, 2015 at 9:09 PM ^
October 12th, 2015 at 9:15 PM ^
Also, thanks op for the cool link
October 12th, 2015 at 9:30 PM ^
Is it able to simulate the number of damaged cars by angry Spartan fans when they leave?
October 12th, 2015 at 9:48 PM ^
October 12th, 2015 at 10:21 PM ^
I haven't trusted that simulator ever since it promised me the Pats were gonna go undefeated in 2007.
October 12th, 2015 at 11:47 PM ^
My simulation :
FINAL SCORE 4 OT -- Michigan (2016): 34 - Michigan State (2016): 30
:O