March Sadness

[SWC75]

This is from a prior post:

Here is a summary of seeding vs. achievement since the seedings began in 1979.
1 seed means you are projected to make the Final Four
2 seed means you are projected to make the Elite 8
3 or 4 seed means you are projected to make the Sweet 16
5, 6, 7, or 8 seed means you are projected to make the Round of 32
Below 8 seed means you are projected to lose in the first round
1979 We were a 4 seed and lost in the Sweet 16 = Even
1980 We were a 1 seed that lost in the Sweet 16 = -2 rounds
1983 We were a 6 seed that lost in the Round of 32 = Even
1984 We were a 3 seed that lost in the Sweet 16 = Even
1985 We were a 7 seed that lost in the Round of 32 = Even
1986 We were a 2 seed that lost in the Round of 32 = -2 rounds
1987 We were a 2 seed that made it to the National Championship game = +2 rounds
1988 We were a 3 seed that lost in the Round of 32 = -1 round
1989 We were a 2 seed that lost in the Elite 8 = Even
1990 We were a 2 seed that lost in the Sweet 16 = -1 round
1991 We were a 2 seed that lost in the Round of 64 = -3 rounds
1992 We were a 6 seed that lost in the Round of 32 = Even
1994 We were a 4 seed that lost in the Sweet 16 = Even
1995 We were a 7 seed that lost in the Round of 32 = Even
1996 We were a 4 seed that made it to the National Championship game = +3 rounds
1998 We were a 5 seed that lost in the Sweet 16 = +1 round
1999 We were an 8 see that lost in the Round of 64 = -1 round
2000 We were a 2 seed that lost in the Sweet 16 = Even
2001 We were a 5 seed that lost in the Round of 32 = Even
2003 We were a 3 seed that won the National Championship = +4 rounds
2004 We were a 5 seed that lost in the Sweet 16 = Even
2005 We were a 4 seed that lost in the Round of 64 = -2 rounds
2006 We were a 5 seed that lost in the Round of 64 = -1 rounds
2009 We were a 3 seed that lost in the Sweet 16 = Even
2010 We were a 1 seed that lost in the Sweet 16 = -2 rounds
2011 We were a 3 seed that lost in the Round of 32 = -1 round
Totals: 26 tournaments. We broke even 12 times. We exceeded our seed 4 times and came up short 10 times. Our net achievement compared to the round we were supposed to lose in was -6 rounds. We were even at the point where we had won the national championship. Of course some of our losses in the “even” years were to lower-seeded teams that had pulled off upsets in earlier rounds so we might still have expected to beat those teams. That was the case in 1979, 1984 and 2004. In none of the four “positive” years did we lose to a lower (worse) seed.

Another way to look at it is this:

From 1977-86:
(In 1977-78 there were no seedings but we lost to lower ranked teams.)
We exceeded our ranking/seeding zero times.
We were even with our seeding in 1979, 1983, 1984 and 1985 but the 1979 and 1984 teams lost to lower seeded teams- we jsut did it in the round we were projected to lose in anyway.
We lost before we were expected to in 1977, 1978, 1980 and 1986.

From 1987-2003
We exceeded our seed in 1987, 1996, 1998 and 2003
We matched our seed in 1989, 1992, 1994, 1995, 2000, 2001. None of the losses were to lower seeded teams.
We lost before we were expected to in 1988, 1990, 1991 and 1999

From 2004 onward
We have not exceeded our seed.
We matched our seeding in 2004 and 2009 but lost to a lower seeded team in 2004.
We lost before we were expected to in 2005, 2006, 2010 and 2011

I think most Syracuse fans have learned to brace themsleves for disappointment based on recent results. Wisconsin is a formidable oppoenent but, even with no Fab, I think we are expected to win this game. Ohio State is a #2 seed but would have been #1 if the Fab announcement had been made before the seedings and I don't think people would be surprized to lose to them. So based on recent results, maybe we should 'expect' to lose to Wisconsin.

Why we were so bad vs. seeding/ranking in 1977-86 and in 2004-2011, i don't know. We were much better in 1987-2003 but 5 even years, 4+ years and 4- years is not an exceptional performance. of course, if you tend to have highly seeded teams, it's harder to exceed your seed and easier to fall short of it. If you are a #1 seed you are expected to win at least 4 games. #2 seeds are supposed to win 3 games. #3-4 seeds are supposed to win 2 games. As we have seen that's not easy to do, even for a very good team.

I just wonder if things are about to change again and maybe we will at elast match our seed this year and in future years or if we will continute to end the season with disappointment. I have no idea what, if anything, controls these things.

 

[mike32768]

Love it - thanks for that.

Statistics are cool, but can be so misleading. I wonder how other elite schools have done?

It is interesting (and correct), that you note it's easier to exceed expectations when the expectations are low. While it seems obvious when that concept is written in word form, looking at just the numbers, it's not so obvious.

Scary part: We only lost once in the Elite-8 and never in the Final-4 rounds... which is caused really by our lack-of-getting-there.
Scarier part: In both years that we were a 1-seed, we lost in the Sweet 16.

 

[RandygoCuse]

That's good data! ESPN just recently did an issue on Cuse and a statisctician noted created something called Performance Against Seed (PASE, I forget what the E was for). Believe it or not Boeheim performed better than average. However, the author also noted that if we take away the runs in '96 and '03, it drops significantly. But I'm not sure it is fair to simply take away two of our best performances. It would be more reasonable to take away our best and our worst. Did anyone else read the article?

 

[cap2nine]

I understand the "projection" based on seeding that you've employed but I'm not too sure it paints an accurate picture. I don't have the time right now to do a thorough analysis by seeding and ratio of success to reach their "projected" rounds but I can say that just over 41% of #1 seeds have made the Final Four so the expectation of #1 seeds reaching the FF may be a little unrealistic when compared to the historical stats.

As a SU fan, my personal expectation matches the team’s goal of a NC. But I think in general a #1 seed is successful if reaching the Elite Eight, or better. Purely based on the historical numbers.

 

[jncuse]

Between 1992 and 2008 we were only a 3 seed or above once. Since that time we have been a top 3 seed each of the last 4 years.

Amazing accomplishment and consistency, but we need to pay it off with at least one run to the final four.

Has any program been a top 3 seed each of the last 4 years? Thinking possibly Kansas, Duke, and Ohio St,

 

[SWC75]

It seems to me simple logic that a #1 seed is expected to reach the Final Four, a #2 seed is expected to reach the Elite 8 and a #3-4 seed is expected to reach the Sweet 16.

 

[KingOtis]

Interesting analysis. I feel the need to point out the following:

1999 We were an 8 see that lost in the Round of 64 = -1 round
2006 We were a 5 seed that lost in the Round of 64 = -1 rounds
2010 We were a 1 seed that lost in the Sweet 16 = -2 rounds
2011 We were a 3 seed that lost in the Round of 32 = -1 round

The '99 game was a toss-up. Last time I checked, the #9 seed has a winning record over the #8. Doesn't feel right losing a point there.

G-Mac was hobbled in 2006. You don't even need to take into consideration that we were over-seeded and they were under-seeded to give that game an asterisk.

With AO out, I don't see how the FF was a realistic expectation. At the very least, I would split the difference and take away one round instead of two.

Marquette already had a win over us in 2011. The committee sabotaged two BE teams by making them play other BE teams way too early. I would call it even.

In addition, you said we matched our seeding in 2004, but we made the Sweet 16 as a #5. So that should be a +1. Maybe you docked us for losing to a lesser seed.

By your estimation, we're a -6, but I think an argument could be made for getting all six of those points/rounds back.

 

[KingOtis]

That's good data! ESPN just recently did an issue on Cuse and a statisctician noted created something called Performance Against Seed (PASE, I forget what the E was for). Believe it or not Boeheim performed better than average. However, the author also noted that if we take away the runs in '96 and '03, it drops significantly. But I'm not sure it is fair to simply take away two of our best performances. It would be more reasonable to take away our best and our worst. Did anyone else read the article?

E = Expectation.

 

[SWC75]

Interesting analysis. I feel the need to point out the following:



The '99 game was a toss-up. Last time I checked, the #9 seed has a winning record over the #8. Doesn't feel right losing a point there.

G-Mac was hobbled in 2006. You don't even need to take into consideration that we were over-seeded and they were under-seeded to give that game an asterisk.

With AO out, I don't see how the FF was a realistic expectation. At the very least, I would split the difference and take away one round instead of two.

Marquette already had a win over us in 2011. The committee sabotaged two BE teams by making them play other BE teams way too early. I would call it even.

In addition, you said we matched our seeding in 2004, but we made the Sweet 16 as a #5. So that should be a +1. Maybe you docked us for losing to a lesser seed.

By your estimation, we're a -6, but I think an argument could be made for getting all six of those points/rounds back.

You are right about 2004. I shouldn't have rated that even by seeding. But we did lose to a lower seed. If you want to hand out asterisks, there could be a lot of them, including K-State not having their second leading scorer and leading rebounder. Of course we've got one in our back pocket this year, as well.

 

[KingOtis]

You are right about 2004. I shouldn't have rated that even by seeding. But we did lose to a lower seed. If you want to hand out asterisks, there could be a lot of them, including K-State not having their second leading scorer and leading rebounder. Of course we've got one in our back pocket this year, as well.

I wouldn't penalize us for beating K-State without Samuels, just as I wouldn't penalize Butler for beating us without AO.

 

[cap2nine]

I don't think you can have it both ways. SWC75's OP is a tool for measuring the success of a team based on its seeding. As I mentioned earlier, I don't feel that the historical results match the projections but it can be used as a consistent measurement nonetheless. If we begin adjusting the data for our benefit due to injuries, matchups and seeding inequities, the measurement loses all meaning. All teams have to deal with the pressures and expectations based on where the Selection Committee assigns them and all teams have to deal with the possibilities and realities of injuries, suspensions and difficult matchups. All I'm saying is, if anyone wishes to apply the "projection by seeding" philosophy, you can't ignore or discredit the actual results.

 

[KingOtis]

All I'm saying is, if anyone wishes to apply the "projection by seeding" philosophy, you can't ignore or discredit the actual results.

No, but you can point out where the results might be flawed. While we have a score of -5 (or -4 if you call it even on the 1999 game), some of those points are forgivable.

That's why stats and formulas aren't the end-all, be-all when it comes to analyzation. For example, Todd Helton puts up monster numbers and we know that a lot of it has to do with Coors Field.

 

[SWC75]

No, but you can point out where the results might be flawed. While we have a score of -5 (or -4 if you call it even on the 1999 game), some of those points are forgivable.

That's why stats and formulas aren't the end-all, be-all when it comes to analyzation. For example, Todd Helton puts up monster numbers and we know that a lot of it has to do with Coors Field.

No stats are the "end all and be all". Their purpose is to put the discussion on an objective basis. You do need to know what the stats are to come to an informed conclusion.

 

[KingOtis]

No stats are the "end all and be all". Their purpose is to put the discussion on an objective basis. You do need to know what the stats are to come to an informed conclusion.

Not disputing that at all. Just so we're clear, I think you did a good job with this and I'm glad someone went through the trouble.

And I would say I'm doing as you suggested. In my view, the informed conclusion is reached when you take the number from your formula and then analyze each game individually to ensure the results are accurate. So instead of -5, maybe it's -3 with a +/- of 2.