So, you're telling me there's a chance (part 3) - UM's chances at the B1G East

Submitted by gpsimms not to… on

You could call this part of a running series started by TaiStreetsMyHero here. Last week, I wrote up a simulation with the various relevant tiebreakers to come up with Michigan's chances at the Big Ten East title.

Honestly, there's not all that much reason for a follow-up post, since a simulation is not needed to calculate Michigan's chances with so few games remaining.  But, the code is already written, and took only a few minutes to update, so here it is:

  Div Champs (out of 10,000,000) %
OSU 9,819,579 98.20
UM 97,624 0.976
MSU 48,565 0.486
PSU 34,232 0.342
Rutgers : ( : (

Alas, Rutgers has fallen out of contention. Also, we can see that last week's results only marginally increased Michigan's chances. While winning our game helped, we still have the two biggest hurdles as ahead, and each of those got more difficult (according to S&P) due to dominating performances by OSU and Wisconsin. Also, Penn State beat Rutgers which was (sad as that is) probably the best chance we had for an upset over Penn State. Finally, the poor performances by Rutgers, Maryland, and Nebraska reduced the chances any of those teams upset MSU or PSU in the final 2 weeks. Relevant games and S&P predictions are pasted below:

Michigan (needs to go 2-0)

Win @ Wisconsin (28%)

AND

Win vs. OSU (31%)

Overall probability of 2-0: 8.68%

Michigan State (needs to go 1-1 or worse)

Lose vs. Maryland (21%)

OR

Lose @ Rutgers (34%)

Overall probability of 1-1 or worse: 47.86%

Penn State (needs to go 1-1 or worse)

Lose vs. Nebraska (11%)

OR

Lose @ Maryland (14%)

Overall probability of 1-1 or worse: 23.46%

Soulfire21

November 13th, 2017 at 3:31 PM ^

If both UM and MSU finish at 7-2, MSU would have the tiebreaker over us via the head-to-head matchup. The only way we get into the BTCG is in a two-way tiebreaker with OSU where we own the head-to-head. Any three- or four-team tie involving PSU or MSU does not work in our favor since we lost to both.

Beat Rutgerland

November 13th, 2017 at 11:29 PM ^

The first tiebreaker is "The records of the three tied teams will be compared against each other."

http://www.bigten.org/sports/m-footbl/archive/081011aaa.html

If that means what it sounds like, then in a 3-way tie between MSU, OSU and Michigan, Michigan would be the only team with only 2 losses and would win outright without going to another tiebreaker.

I'm pretty sure we only need to win out (ha, only) and have PSU lose.

Muttley

November 14th, 2017 at 12:32 AM ^

They should start with

-1. If at any point in the procedure below, a team(s) is eliminated, start again at step 0 with the remaining teams under consideration

0. The B1G records of all teams not yet eliminated will be compared

1. The records of all remaining teams in games against each other will be compared

...

They could use a lot fewer words to say the same thing.  No need to have separate cases for two teams, three teams, etc.

Soulfire21

November 13th, 2017 at 3:38 PM ^

The first tiebreaker is record vs. each other for the tied teams. If UM, OSU, and MSU are all 7-2 we'd have OSU at 1-1 (win over MSU, loss to UM); MSU 1-1 (win over UM, loss to OSU); and Michigan would be 1-1 (win over OSU, loss to MSU).

It's a draw.

The next tiebreaker is intra-divisional record. UM would be 4-2 in the division, OSU would be 5-1, and MSU would be 5-1. That drops us out of contention, and the tiebreaker becomes the head-to-head winner of OSU and MSU, which was OSU, so they go.

Soulfire21

November 13th, 2017 at 3:41 PM ^

If MSU and PSU each pick up at least one more loss, both would have 3 conference losses. Assuming Michigan wins out, that's only 2 conference losses. Michigan and OSU are tied atop the division at 7-2. Because this scenario requires a Michigan win over OSU, Michigan represents the East by virtue of the head-to-head result against OSU.

gpsimms not to…

November 13th, 2017 at 4:55 PM ^

but, when TaiStreetsMyHero posted this calculation the first time four weeks ago, our probability of winning the division was like 0.00062%, or something crazy like that.

Our chances of winning the division have increased about 1500x since then, so I have decided these posts are "good luck."

If we're not mathematically eliminated next week, expect another post.

Hail-Storm

November 14th, 2017 at 11:41 AM ^

What would be the point of not sharing the probabity of Michigan winning the B10 East? It's nice to know there is a shot, and, like you said, it's nice to trace the percentage chance on the rise to get there.

Of course there is probably like a 99% chance we won't go (someone would have to do the math on that ;), but it's fun to see that there's still a chance.  Keeps even more meaning in the last few games to win.

KennyHiggins

November 13th, 2017 at 3:39 PM ^

Then we can start paying attention to the statistical analysis for the final weekend.  Give me wins the next TWO Saturdays, and honestly, it's a great season, with or without the B!G title.

sum1valiant

November 13th, 2017 at 7:30 PM ^

This. We go 2-0 the next two weeks, finish the regular season at 9-2 with two top-10 wins to finish the season, and beat a top 15 team to carry the momentum into the off-season, I'm on cloud 9 heading into next year returning basically the entire team.
Get beat up twice, and it's going to be a looooong nine months trying to figure out how to right the ship.

uncleFred

November 13th, 2017 at 9:24 PM ^

Michigan is 8-2. If they win out they'll be 10-2 in the regular season. But if they end up 8-4 it will NOT be a "looooong nine months trying to figure out how to right the ship". You right the ship by not losing your starting QB early in game four of the season. 

Come on. The consensus here when the season began was that, without injuries, 8-4 or 9-3 would be a pretty good rebuilding year. If you look at the injury list at this point 8-4, which is the least regular season possible, should be cause for deep satisfaction.

sum1valiant

November 13th, 2017 at 9:44 PM ^

My apologies- made a mistake on the schedule. Although I didn't apply any math whatsoever, just absent mindedly assumed an 11 game schedule. Since we're being condescending dicks though, I'll clue you into the fact that there was never a "consensus" on a 4 loss regular season that included losses to MSU, OSU, PSU, and Wisconsin.
"The injury list" that you refer to is eerily similar to every other football team's injury list. People get hurt playing football, and rosters are particularly depleted after ten games (thanks for the correction). Anyone that points to said injuries will be grabbing at straws in an attempt to justify a season where we didn't beat a team with a winning record.

TrueBlue2003

November 14th, 2017 at 1:37 AM ^

if we lose these last two.  This was always going to be a rebuilding year after we sent like 14 guys to the NFL. We've endured injuries to our starting QB and best receiver.  We were projected to win like 8.5 games by vegas and the fancystats even before those injuries.

With almost everyone coming back, I'll save you nine months of pondering how we win the conference next year: find a couple OTs.  That's it.  It will be that simple.

If we don't, we'll be an 8-9 win team again, which isn't great but it'll do on the way to 2019, which should be really, really good.  Especially if Gary comes back.

MGoBeast

November 13th, 2017 at 3:44 PM ^

Maybe they beat 1 of those two, but I'm too lazy to google. At this point, they have (2) wet fart games to end the year and no motivation to win out since they just got jail-sexed by OSU. I wouldn't be suprised to see them lost one of those. Not that I think it's likely either. I think Lewerke gives them enough to win.