2010 Michigan Prediction and Probable Season Expectations
We had over 180 participants in the "Prediction Survey". The results were a bit cheery, at 9-3. Perhaps everyone was a bit geeked up with the "Countdown to Kickoff" starting that day and various other prediction threads and hype videos. Without futher ado, here's the data...
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
|
||||||||||||||||||||||
At one point within the predictions thread, another user thought it'd be a good idea to provide a probability or likelyhood for a Michigan victory in each game. The online survey maker I used wouldn't tally the results in any valuable way, so I had to transfer all the data into an excel spreadsheet. I've uploaded the individual game results onto a Google Docs page. The link to that document is here:
Here are the results of the probability survey
Interesting, yet not entirely unexpected, the results came out to an 8-4 finish. What was surprising was the one game that flipped was the Penn State game. I was more expecting Notre Dame, or even perhaps the Michigan State game to be closer. Anyway, thanks for everyone's involvement. We had over 130 participants in the probability survey.
But I'm a little confused. You asked users for their predictions and then for their probabilities? Not sure what the difference is.
I guess with a prediction it's either win or loss, but in a probability survey, everyone gives their percentage chance of winning each game and then you averaged them? Is that right?
Yes, the numbers you see above were the average "confidence level" for a Michigan win. Obvsiously, above 50% translates to a UM win.
You said that PSU was flipped, but don't both studies show the MGoCommunity is chalking that up as a loss?
The former is a toss-up. I wouldn't say that 91-90 is indicative of MGoCommunity suggesting PSU is a loss, but that MGC is genuinely torn over the outcome. That's practically a tie.
Yeah, I guess it is, now that I look at it again. Last night, when some results were being discussed in the thread, the participants had UM as a winner in that game. We were at 150 some particpants. The numbers above show just over 180 participants, so overnight and into this morning, 30 some odd people took the survery. I never looked at it that closely again. Good catch.
EDIT: I just went back through and removed any forms that didn't include usernames. There were quite a few "anonymous" forms that had UM going 0-12. With those gone, the PSU game is at 50.57% for a UM win.
August 6th, 2010 at 11:06 PM ^
Did you also delete all the respondents that put us at 12-0?
I honestly don't think there were any. I'll check right now.
EDIT: There was one
delete the UMass/BGSU loss guy also. He probably had us going 0-12 given those predictions...
The probability survey was a good idea, paints a more accurate picture. I'm also a little surprised by the pessimism toward the PSU game.
You could then take these percentages and create probabilities for each number of wins. That would be pretty cool to see being that a 47% "chance" of losing to Penn State is still pretty much 50/50.
I'd be especially interested in the "probability" that Michigan goes undefeated. Would you just multiply all the win percentages together? I haven't taken stats since high school...
I put quotes around certain words because it's not really a probability of course. Just our confidence level.
Looks like both methods came up with 8-4. The survey had us losing Iowa, PSU, Wisc, and OSU. Same as the probability method. I would be satisfied and hungry for 2011 with 8-4.
August 8th, 2010 at 10:42 AM ^
Me too. I like this team and am very excited to see what they have. I am tired of everyone picking on Michigan and Michigan football too. 8-4 and a happening offense would be fine as a stepping stone.
1997 and 2006 followed 8-4 and 7-5 seasons, but each game in the prior season could have been won.
If we go 7-5 and lose five nailbiters I'm OK with that. If we go 8-4, win all of the nailbiters, but lose four 38-13 contests, then there's not much reason to be stoked for 2011.
Whose the one troll that had us losing to UMass and Bowling Green!?
because we live under da bridge?
But seriously, someone thinks Michigan will lose to UMass and Bowling Green? That's a h8r!
If you treat these numbers like probabilities, then you can calculate the probabilities of the final season record. Here's what I get:
5-7 = 0.6%
6-6 = 3.8%
7-5 = 14.3%
8-4 = 29.2%
9-3 = 30.9%
10-2 = 16.7%
11-1 = 4.1%
12-0 = 0.3%
A pretty optimistic group in my estimation...I like it!
0-8 = 0.1%
1-7 = 0.9%
2-6 = 5.0%
3-5 = 15.4%
4-4 = 27.1%
5-3 = 28.3%
6-2 = 17.0%
7-1 = 5.5%
8-0 = 0.7%
I used the top set of numbers for the above calcs instead of the 2nd set. Here's using the correct numbers:
3-9 = 0.2%
4-8 = 1.3%
5-7 = 5.1%
6-6 = 13.1%
7-5 = 22.6%
8-4 = 25.9%
9-3 = 19.6%
10-2 = 9.3%
11-1 = 2.5%
12-0 = 0.3%
...certainly think alike. I've been feeling 8-4 or 9-3 for some time now. Would love to see 10-2 with losses only to Iowa and Wisconsin!
I have a question about Penn State: did they lose anything of value on defense and, if so, how much do those losses affect them?
Surprised at the Iowa results - we almost beat them on their turf last year. Maybe I'm getting carried away with enthusiasm but I like that game.
August 6th, 2010 at 11:01 PM ^
Spot on
understand the power of triple revenge?
The only team this season that probably has a decisive talent advantage over us is OSU. We have a chance against every other team. PSU - OL and QB questionable, IA - we will score on their defense, WI - your typical big bruising team...but stop the run and you could win.
I'm sure there will be several games that will come down to the wire where a mistake on our part will result in a loss and a mistake by the opposition will be a win. So 7-5 or 8-4 is looking good unless we have a bad case of the injury bug!
Comments