Ohio State is no Longer #1 in All-Time Winning Percentage
After the loss to Missouri, the former #1 team in all-time winning percentage now sits at #2 on the list. According to Winspedia (which does not have the bowl game loss listed yet at the time of posting this)
Current Rankings
- Alabama - 965-336-43 - 73.40%
- Ohio State - 964-333-53 - 73.37%
- Michigan - 1002-353-36 - 73.33%
- Notre Dame - 948-338-42 - 72.96%
For those wondering. Due to the number of games played not being equal. Yes, we will pass Bama with a win, but we can not pass Ohio State this season to regain the #1 spot on this list. I didn't plug in any other results because all I care about is 15-0 - 100%!
December 31st, 2023 at 9:46 PM ^
Those percentages might reflect what the NCAA officially recognizes as win %, but as with virtually everything, they corrupt it. They're counting ties as half a win, which they are not.
If you divide the total number of wins by the total number of games played, you will see that Michigan has won over 72%, while Bama is at 71.8 and OSU at 71.4.
I don't know about you, but it seems pretty obvious to me that actual win percentage is what percent of the time a team wins when playing games of football.
OSU has never been #1 by this measure, nor Bama
December 31st, 2023 at 9:56 PM ^
I guess. But mainly I just it find it weird that Michigan has so many fewer ties. Especially because I remember 1992.
December 31st, 2023 at 10:00 PM ^
Username checks out
December 31st, 2023 at 10:09 PM ^
Hi there!
Former SID here chiming in to say this is incorrect.
The accepted way to calculate winning percentage when ties are involved is to in fact count them as half a win and half a tie.
And it's not just how the NCAA does it, it's how the NFL does as well.
December 31st, 2023 at 10:36 PM ^
It's not a win though
(I believe you that's the system, just seems like a dumb system)
December 31st, 2023 at 10:50 PM ^
Its also not a loss though, and has to count as something.
For example, last year in the NFL, the Commanders finished 8-8-1 and the Packers finished 8-9. So Washington finished ahead of Green Bay, because their win% was .500 and GB was .471.
The Packers won 47% of their games and lost 53% of them. If you just say winning percentage is what Asquaredroot wants, then the Commies won 47% of their games....but they did not lose 53% of their games like GB did. Their "losing percentage" was also only 47%. So you have 6% unaccounted for. Ties counting as .5 accounts for the missing percentages without overly benefiting or punishing a team.
And like I said, the NFL also does this, so it's not the "NCAA corrupting it" like Asquaredroot says. It's just how statistically you handle ties if you dont have a points system like hockey or soccer. I'm all about calling out the NCAA on corruption, but thats not what it is here.
December 31st, 2023 at 10:54 PM ^
Thanks Quailman. I don't know what SID stands for. Internet searching SID in the context of statistics didn't shed any light after 30 seconds or so of effort, and I assume you aren't insulting yourself with that acronym which rules out some of the results. 😁
It makes sense that both the NFL and NCAA would value a tie more than a loss, but if I understand English correctly, they're not doing the math right.
Or conversely, they aren't using English correctly for the math applied.
That's more like a 'points per game' percentage where a win = 1 and a tie = 0.5.
Or call it a Win-Loss-Tie percentage.
The main point is, Michigan has won a higher percentage of the games they've played than any other team on the list.
December 31st, 2023 at 10:56 PM ^
sports information director
December 31st, 2023 at 11:04 PM ^
Thanks. Ill leave my proclamation of SID ignorance unedited.
To clear up confusion for those of us who dabble in both accurate math and English, I recommend they change the term to 'Win-Loss-Tie percentage' for the stat and add a second column for 'Win percentage' at least when it comes to all time stats.
December 31st, 2023 at 11:03 PM ^
Please see my explanation above.
It's not like a "points per game percentage," its how you account for ties when you dont have a points system like hockey or soccer. You have to account for the ties statistically because winning percentage and "losing percentage" have to equal 100.
They are using English correctly, and they are doing the Math right. You are either being a bit too on the nose/pedantic with what you want winning percentage to mean, or just want to make a winning percentage where UM is #1 already.
December 31st, 2023 at 11:12 PM ^
Call it pedantic or whatever judgement you like.
The statistic is inaccurate as named when it comes to all time stats and is irrelevant for season stats since a point system like the NHL uses would suffice.
Yes, I do want UM to be #1 all time like many on this blog. Doesn't change the fact that the stat calculation/naming is bogus.
December 31st, 2023 at 11:53 PM ^
I'm not judging. Im just aware that you are dug in with what you want it to be, when thats just not what it is.
The statistic is not inaccurate as named, you just dont like how its named. Also, its not irrelevant for season stats, see my example from the NFL above.
January 1st, 2024 at 12:27 AM ^
While we're working at being technically correct, I have a problem with the NHL and NFL "percentage" columns. They call them percentages but they list them as decimal fractions, .663 or .571. If they were percentages they would shift the decimal point over two places, 66.3% or 57.1%.
I read the numbers and it makes me want to axe someone a question about what they seen.
January 1st, 2024 at 1:44 AM ^
I don't watch rooting for my team to not lose the most compared to another team. I root for them to win more than the other team.
January 1st, 2024 at 2:42 AM ^
It's true. I'm dug in that win % should reflect the % of times a team wins games it's played.
It's a simple concept regardless of which team one is rooting for.
January 1st, 2024 at 1:49 AM ^
“You have to account for the ties statistically because winning percentage and "losing percentage" have to equal 100. “
No they don’t. If there are three possible outcomes for each game then winning percentage, losing percentage, and tying percentage must all sum to 100%. But there is no a priori rule that you must only use winning and losing percentages to deal with a sport that allows win, loss, and tie as a result for each game.
January 1st, 2024 at 12:06 PM ^
Actually there is no objective reason win and loss percentages have to equal 100. Mathematically that's only true if they account for all games, and in all of these cases they don't.
Like Asquared, I acknowledge that the convention is to give half a credit for a tie, and it makes sense for an organization to do that when measuring one team's performance against another. Most would agree that a tie is more successful than a loss.
But it's only a convention, and it doesn't change the fact that Asquared's calculation is a more precise measure of winning percentage, and "winning percentage" more precisely describes what he is calculating. It would also be just as valid if that were the convention.
January 1st, 2024 at 9:03 AM ^
Accepted? I don't accept that. So what if the NFL accepts that. After watching the Dallas-Detroit game do you think the NFL knows what it is doing?
Winning percentage is wins/games played.
That's what I accept.
December 31st, 2023 at 10:11 PM ^
As the saying goes, ties are like kissing your sister, so of course Bama counts them.
December 31st, 2023 at 10:45 PM ^
Well, she is a good kisser
January 1st, 2024 at 10:46 AM ^
Was wondering how they count ties. IMO a tie is a loss for both teams and should be counted as such. Anyway, thanks for clarifying
December 31st, 2023 at 10:15 PM ^
Looking at your post and the numbers, visually it looks like Ohio loses 4 basis points with a single loss. To pass Alabama after the Rose Bowl, we have to gain a net of seven basis points on Alabama. Since we’ve played more games that could be roughly -4 basis points per Alabama and +3 for us. That would be consistent with your post
But that would leave us 1 basis point behind Ohio and maybe a fraction of a basis point ahead of Alabama.
So in that case, how do we not pass Ohio if we win the championship?
December 31st, 2023 at 10:37 PM ^
Ohio State is 964-333-53 for a .7337
If UM beats Bama,
Bama falls to 965-337-43 for a .73345 and UM improves to 1003-353-36 and .7335
If UM wins the title, they are now 1004-353-36 and .7336
January 1st, 2024 at 11:24 AM ^
Helpful stats, but if you're only taking the win-plus-half-ties percent of a national champ Michigan to four decimal places, you should round up to .7337. Not saying they would tie OSU ... the nice man below gave the more precise picture. Just pointing this out since we're being sticklers for numbers.
January 1st, 2024 at 1:48 AM ^
Our win percentage would be .73367, and OSU's would still be .73370
December 31st, 2023 at 10:18 PM ^
Hello.
Fuck Notre Dame. Also have a good night.
December 31st, 2023 at 10:35 PM ^
is this the actual top 4/mount rushmore of cfb? no arguement here. i would guess usc as 5 without looking at more data. We should claim George's spot tonight. HnY
December 31st, 2023 at 10:54 PM ^
The only thing that matters is win total. (Unless we take back the win % spot also)
January 1st, 2024 at 4:00 AM ^
No matter.
January 1st, 2024 at 6:40 AM ^
Two words. Catapult cheaters.
January 1st, 2024 at 10:11 AM ^
Doesn't it look nice to be the only Team with at least 1,000 wins?!!!