Should preseason predictions be used to determine success?

College Football27

I know there is a lot of angst after the Clemson game and many people are upset we lost the game. The question I have been asking myself is “Should we have expected to win that game?” Or to extend it further “Are we expecting too much by wanting to win the division?”

The argument is, since we were predicted at 4th place, finishing higher than that is considered success. This argument relies on believing the preseason predictions are an accurate portrayal of a team’s and its foe’s abilities. So we need to determine how accurate the predictions verses results have been in the past. If the predictions are accurate then the argument is valid.

To test this I went back to find the preseason and the final results back to 2005. Here is the chart for that data (note 2010 is not final data)

The Coastal division is pretty easy as VT is at the top and Duke is at the bottom and since we are not in that division I will not dwell on that.

The Atlantic seems to be much more difficult to predict.

Here is the chart on how often the predicted finish is correct for the Atlantic division(2010 information is not included as it is not final)

Conclusions from this chart

1)      It is much easier to pick the bad teams than the good. Only three times out of 10 did the bottom two finish higher than predicted.

2)      It is hard to pick the top two. 6 out of ten times the teams picked to finish in the top two finished lower than the top two.

3)      The teams picked in the middle of the pack had a 40% chance of ending up in the top 2.

4)      The winner of the division came from the top two 3 out of the 5 times.

What does that mean? It means preseason predictions are wrong far more often than they are right and indicates using preseason predictions to determine success is not very accurate. Once the season is underway then the preseason expectations lose importance because they are made on inexact information and assumptions.

Other random observations

1)      The fourth place team never finished below fourth and 80% of the time finished higher than fourth.

2)      NCSU has dominated the bottom two slots for the past four years.

3)      The third place team never finished third. 60% of the time it finished higher.

4)      4 different schools have won the division in the last 5 years

About Rick

1992 and 2002 graduate from NCSU. Born and raised an NCSU fan. I remember the good ol' days and they weren't in the last 20 years.

College Football

27 Responses to Should preseason predictions be used to determine success?

  1. wolfbuff 11/09/2010 at 2:32 PM #

    This confirms that it’s difficult to predict the outcome of a 12 game season in advance (PS3’s season simulation notwithstanding). But to answer your question, we as humans “learn forward”. We change our expectations based on recent experiences or events. Given that we were 6-2 and coming off a big win against FSU, yes, I think we should reasonably have expected to win that game LAST WEEK. If you had asked almost anyone before the season started, I would think most people would have said no. But that was based on the past 3 years.

    Another interesting question to pose would be “how much do the preseason rankings determine the national champion”? I know SOS has something to do with that now, but if teams A and B are ranked higher than C at the beginning of the season, and all 3 finish undefeated, it’s almost impossible for team C to get to the national championship – especially if they started out high in the rankings, or not ranked at all. That’s why I never thought there should be any rankings until mid-season. I know they don’t come out with BCS rankings until mid-season, but that is based partly on rankings – which have been going since preseason.

  2. Daily Update 11/09/2010 at 2:46 PM #

    Why should we reasonably expect to win the Clemson game when the point spread had us as underdogs? It was reasonable to expect us to have a chance to win based on the size of the spread, but it was certainly not unreasonable that we lost to Clemson.

    -We are 6-2 against the spread this year with 3 outright wins as underdogs(UCF, GT, FSU) and only 1 loss as a favorite(ECU).

    -We are 2-1 as road underdogs

    -We have been underdogs in 5 of 8 FBS games(UCF, VT, FSU, GT, and Clemson having gone 3-2.) Even as the season has progressed, the non-emotional betting public has still has us as underdogs against FSU and Clemson. Hell, we were an underdog against VT when they were 0-2 and had just lost to JMU.

    So as we have won and appeared better, the lines have gotten closer, but still not to the point that our expected win total by the betting public should have been 8+ wins.

    -We will likely be underdogs against UNC and Maryland as well(they are just as good as us). Maybe not Maryland depending on how they do vs. UVA and FSU.

  3. tjfoose1 11/09/2010 at 2:47 PM #

    “Should preseason predictions be used to determine success?”

    No

    “Should we have expected to win that game?”

    Yes. Vegas, the media, or the public at large should never set our expectations for us.

    We don’t accept it for basketball, we shouldn’t accept it for football.

    We should have expected to win vs Clemson because we are a better team and there were match ups that should have been exploited to enable us to win. Clemson may have better talent, but we are (or at least were) the better team. The coaches failed, that is why we lost.

    On the other hand, we shouldn’t have ‘expected’ to win vs FSU, and we shouldn’t ‘expect’ to win vs UNC.

  4. ChemE79a 11/09/2010 at 3:18 PM #

    I believe “expect” is being used in two different senses here.

    One sense is that there is a mathamatical liklihood that we will win the game. Arguements based on betting lines and the like use that definition.

    The other sense is that “we” as a fan base or a team believe that the NCSU football team is good enough to win the game. Arguements based on pain of loss or lost opportunity use that definition.

    Unless folks settle on a definition there will be no agreement.

  5. Rick 11/09/2010 at 3:50 PM #

    “We are 6-2 against the spread this year with 3 outright wins as underdogs(UCF, GT, FSU) and only 1 loss as a favorite(ECU).”

    A spread is not a predictor of who will win as much as it is an attempt to even the betting so everyone bets equally on the two teams. Therefor perception comes into play. IMO the betting line is not the be all, end all for deciding who should win. It certainly should not be used to justify our poor performances by saying “well Vegas expected us to lose”.

  6. Daily Update 11/09/2010 at 4:05 PM #

    ^I understand exactly what the spread is. Asking fans of their own teams is certainly not the be all, end all for deciding who should win either.

    Is there any game we shouldn’t win?

  7. Rick 11/09/2010 at 4:14 PM #

    “s there any game we shouldn’t win?”

    Saying we should have won the Clemson game because Vegas said so is misleading and you know it is.

    Of course that was beside the point of the article.

    The point is to say that we beat expectations is not a legitimate argument.

  8. Daily Update 11/09/2010 at 4:38 PM #

    Using the spread to see who is going to win isn’t necessarily misleading. In some situations, it could be.

    The spread gives you a good idea of the how the matchup is perceived by people only interested in winning money…not because they pull for one of the teams involved. The larger the spread, then the more often the favored team wins the game. The closer the spread, the more likely an underdog wins the game because the game is more of a coin flip.

    I didn’t say Clemson should have won. I said Clemson winning wasn’t unreasonable going into the game or based on how the game was actually played(yardage totals even, TO margin with 1, penalties fairly close, etc…).

    It was a virtual toss up.

  9. Rick 11/09/2010 at 4:48 PM #

    I guess I do not understand taking expectations and applying them to past events. Once the event occurs the expectations are irrelevant.

  10. wvillepack 11/09/2010 at 5:04 PM #

    At the beginning of the season I expected us to go 5-7. After starting 4-0 I expected to go 9-3 or 8-4. Now expect us to go 7-5. Expectations change after we see the team (and other teams) play each week.

  11. BJD95 11/09/2010 at 5:18 PM #

    The coaching staff is responsible for the health of the program in Season Four, period. Beating pre-season media predictions is completely irrelevant. That just gives the staff credit for the hole they have dug in years previous.

    I don’t see how anybody could consider anything less than 7-5 at all successful. And yes, 7-5 would feel hollow given the chances we have blown. And no, I can’t get past the fact that we didn’t give ourselves a reasonable chance to win on that fateful 4th and 1.

  12. MatSci94 11/09/2010 at 5:24 PM #

    Should preseason prediction count in success?

    Absolutely. We are basically in line with ‘absolute best case’ for our predictions before the season. Credit has to be given for that.

    Is it the only factor?

    You would be a moron to only consider that. What I think is maddening for us is that we are really only a handful of plays from being unbeaten, and that makes us think about what could have been. Ironicly, I think the result of the Clemson/VT games are that we feel worse about where we are than if we had just been soundly beaten.

  13. VaWolf82 11/09/2010 at 5:25 PM #

    to say that we beat expectations is not a legitimate argument.

    Sure it is. When you take over a bad team/program, that’s the only way to fairly evaluate the coaching staff. Howeer at some point in time, you have to switch your evaluations from “expectations” to “wins”.

    The only discussion worth having is when do you make the transition.

  14. BJD95 11/09/2010 at 6:52 PM #

    Season Four is always the “shit or get off the pot” year for me.

  15. packplantpath 11/09/2010 at 8:31 PM #

    I agree BJD. I’ve been a, maybe not supporter, but not a critic of the last few seasons. This was his year to show me what he can do with a team that is mostly his.

    I reserve judgement until it’s over, but I have been both pleasantly surprised and disappointed depending on the game.

  16. Pack Mentality 11/09/2010 at 8:46 PM #

    My preseason expectations were for us to be poor to mediocre. We still have a chance to meet or exceed my expectations. A losing record is poor and we could still end at 6-7. A record of 7-6 is mediocre. An 8 win season or better or a season that includes 14 games is good. An ACC title is a once in a generation form of greatness for State.

  17. albunde6 11/09/2010 at 9:27 PM #

    Preseason expectation should be one of the measures for success or failure of a team. Those expectations can change due to teams performance, injury, competitions performance.

    My expectation was for 7 wins. I said and stand by the prediction, “if the offense could average 38 points per game they would not loose more than two games”. I also predicted that Dana Bible was the only one who could hold this team under 38 point average.

    I feel this team is a success, but this team has disappointed.

  18. El Lobo Loco! 11/10/2010 at 5:27 AM #

    You have touched on the definition of failure by success.

    Initial expectations were low (4th), then they crept sky high (BCS Champs) to be returned to a more normal position (2nd in division?)…. thus a failure

    I believe this team IS a success, and should be the base for a higher expectation next year.
    That does not erase the disappointment in the bad decision that lead to losses.

  19. Rick 11/10/2010 at 8:25 AM #

    “Absolutely. We are basically in line with ‘absolute best case’ for our predictions before the season. Credit has to be given for that.”

    But if those expectations are almost always wrong (and they are) then why do the predictions matter? If the predictions are inherently wrong then it is a ridiculous comparison.

  20. Rick 11/10/2010 at 8:28 AM #

    “Sure it is. When you take over a bad team/program, that’s the only way to fairly evaluate the coaching staff. Howeer at some point in time, you have to switch your evaluations from “expectations” to “wins”.”

    Then by the expectations standard TOB has failed miserably.
    In his three years he has been picked to finish 3rd, last and last
    He has finished next to last, last and last.

    He has not met expectations once and met the expectation of being the worst team the other two times.

  21. VaWolf82 11/10/2010 at 10:18 AM #

    But if those expectations are almost always wrong (and they are) then why do the predictions matter?

    I see a major difference between media predictions and expectations of the fan base. Media predictions and the fact that they are usually wrong has no impact on me or my expectations.

    TOB took over a 3 win team and in his first two years won 5 and then 6 games to make a bowl game. Last year was disappointing, but it is impossible to say how much of that was due to the injuries. (Blaming it all on injuries is just as silly as ignoring the injuries.) So on balance, TOB has met and/or exceeded my expectations in two years out of his first three.

    The other part of expectations that your year-by-year look skips over completely is how bad State was when TOB took over and how long it was going to take to turn the program around. Both parts are subjective judgement calls and will vary (probably widely).

  22. Rick 11/10/2010 at 11:14 AM #

    I see a major difference between media predictions and expectations of the fan base”

    Granted but there is no way to quantify fan expectations. However, I think the same concept holds true. Fan expectations set in preseason should and do change.

    “The other part of expectations that your year-by-year look skips over completely is how bad State was when TOB took over”
    While I agree we were bad there is no way to determine how bad.

  23. Conrad 11/10/2010 at 11:48 AM #

    Hey guys this articIe is taIking about ACC fbaII and preseason conference ranking. And if i am not mistaken. State was predicted to finish 4 th in the atIantic division. But i was just debating another WoIfpack fan about this very same points in another State chat room.

    So i ask you guys this, how many of you thought that this team wouId be in this position at this time with aII of their concerns coming into the season ? Nobody did. And i understand it. Remember heading into the fbaII these were TOB concerns. 1. Mike Archer and his very young defense with question marks about wouId Irvin be his oId seIf. WeII Irvin has and aIso Archer too with his conservative defensive caIIing. Whereas his corners are setting about 10 yrs to protect from the big pIays from the WRs. 2. A very un and inexperience offense Iine with very IittIe to none game experience. 3. This is with no 5 stars pIayers, but with some 4s,3s,2s,1s star pIayers.

  24. VaWolf82 11/10/2010 at 12:01 PM #

    While I agree we were bad there is no way to determine how bad.

    Yes, but we can still gather some data to get a general indication…which I did when I documented the 2007 roster and their scout ranking coming out of high school:

    http://www.statefansnation.com/index.php/archives/2007/10/09/2007-football-roster/

    The bottom line for me was that State was in really bad shape when Amato left….far worse than State fans wanted to admit.

  25. TheCOWDOG 11/10/2010 at 1:34 PM #

    The true opening line in Vegas was +2.5. It is well known that odds makers award 3 pts. in their handicapping formula to the home field advantage. State was a 1.5 pt. favorite if on a nuetral field. To me therefore, the better team and should have won that game.

1 2

Leave a Reply