| 0 |
Subject: How to Determine a Team's Greatness?
Posted by: KrazyKoalaBears
- Donor [266182910] Sat, Mar 23, 2002, 12:44
Ok, here's your chance to sound off. I'm working on a game similar to the TSN/SW model, except in my version you buy and sell teams and a team's price is based on their performance, not buys/sells. The idea originated in the Hockey Forum during this season, but I didn't have time to build anything for Hockey and/or Baseball, so I'm looking to Football. The idea is simple. You have 5 slots to buy teams. You have 2 AFC slots for 2 different teams, 2 NFC slots for 2 different teams, and 1 WildCard slot where you can buy any team you don't already own (so no duplications). Initial prices are based on last years record/stats and change according to the current year's record/stats so that price changes can actually be calculated ahead of time and so that performance on the field is rewarded with a price increase or peanalized with a price decrease. Further, there will be points, which will be the main goal of the game, based on team performance and that's where the question of dominance comes in. What 3-5 factors do you look at when considering a teams dominance or greatness? Points? Yardage? First Downs? Point Differential? Turnover Margin? Number of Gatorade Cups? So here's your chance to say what you think determines a great team. Speak up and let the debates begin! NOTE: Try to rank your factors in the order of importance to provide as much info as possible. |
| 1 | Sludge
Sustainer
ID: 113368
Sat, Mar 23, 2002, 13:29
|
Obviously, if you're doing this on a game-by-game basis, #1 has to be whether they won or lost.
I'm not so sure that yards gained or yards allowed is the best indicator. A crappy team can pile up a ton of yards if they're in catchup mode all the time. I would seperate out rushing yardage and passing yardage. Ditto for rya and pya.
Kick return yards is just out of the question. That just rewards teams with crappy defenses.
Turnover margin would be a decent measure. As far as points go, I would reward a team seperately for points scored and points allowed. Look at the Ravens from a couple of years ago. (Of course, depending how you make the scoring system based on points, it could end up being equivalent to rewarding on point differential.)
There's my initial, incomplete, thoughts on the matter.
|
| 2 | TBRaiders
ID: 5816109
Sat, Mar 23, 2002, 15:47
|
I agree with Sludge on Win/Loss being the first and most important factor.
The only other two factors I think you need are offensive and defensive efficiency. Both could be based on a scoring model for Yards, Points, and Turnovers.
You could add twists to the game like bonus points for all five teams winning, or add extra points for the highest team efficiency based on your scoring model.
TB
|
| 3 | smallwhirled
Donor
ID: 29232811
Sat, Mar 23, 2002, 20:24
|
W/L record is a given, and I feel that point differential would be the second most important statistic in determining greatness. The third stat IMHO would have to be the time of possession. Dominance in ball control usually translates into domination of the game, and to be great, you have to be dominant.
smallwhirled
|
| 4 | Sludge
Sustainer
ID: 113368
Sun, Mar 24, 2002, 01:46
|
Unfortunately, smallwhirled, time of possession is often meaningless. A high-octane offense will generally spend far less time on the team (e.g. the Rams) than a grind-em-out offense. Both can be every bit as dominant, one just takes a lot less time.
|
| 5 | smallwhirled
Donor
ID: 29232811
Sun, Mar 24, 2002, 11:27
|
Yes, for the Rams, but this is not college football where the college teams from Florida dominate teams with 2 minute quick strikes. With the exception to the Rams, I feel that TOP is pretty important. TOP will not decide every game, but a team that has a strong TOP will probably have a respectable record. I don't know, Sludge, but having the ball the whole game just rips at an opponents heart, and I feel that it is almost as important as turnover margin (in the long run). Yes a team with a -4 margin in one game will probably not win, but there were some very good teams (Rams) with a horrible turnover margin.
|
| 8 | Sludge
Sustainer
ID: 113368
Sun, Mar 24, 2002, 12:33
|
Basically, it all comes down to the basic question:
How strong of a correlate is [insert statistic here] with winning?
A secondary question is:
Is statistic X basically measuring the same thing as statistic Y? If so, then we probably only need to use X or Y, but not both.
|
| 9 | smallwhirled
Donor
ID: 29232811
Sun, Mar 24, 2002, 17:00
|
Sludge, I haven't looked for the "X" yet, but I totally agree with your previous post.
|
| 10 | Sludge
Sustainer
ID: 54131712
Mon, Mar 25, 2002, 12:11
|
At any rate...
KKB - If you can dig up data on past performances, I'd be more than happy to donate a little time to analyze which statistics are more closely associated with winning. Given the nature of the data, however, I'd like to have at least 5+ years. Best case is a game-by-game breakdown including opponent's statistics (for things like turnover margin, point difference, etc.)
|
| 11 | Toral
Sustainer
ID: 2111201313
Mon, Mar 25, 2002, 12:20
|
Just saw this thread. Like Sludge said, correlation with winning is what would count. Palmer and Thorn's The Hidden Game Of Football looks into this topic, as did a short-lived series of annuals called "Football By The Numbers". Their consclusions were similar. As I remember what they found, lots of things you would expect to corellate with winning don't. Yards is a bad factor to use as a determinant because very good teams can have lots of yards piled up in the air against them. Likewise time-of-possession means nothing -- the Bills running the K-Gun could dominate a game and lose TOP 35-25. An effective ground game comes up looking very unimportant.
Best single stat is probably difference between avg YPP on offence and defence (if you consider that a single stat). The one truly "super stat" considers turnovers about -40 yds and then incorportaes them in ydg figures, then comes up with YPP figs as above.
Toral
|
| 12 | sarge33rd
ID: 46231913
Mon, Mar 25, 2002, 13:01
|
as has been stated prior W-L means everything. Secondly, (though seemingly repetitive) W-L on the road. The great teams find ways to win more often than not, away from home.
3rd, is an unofficial stat I used to keep back when I bet heavily on the NFL. YPPS and YAPPA. Yards per point scored and yards allowed per point allowed. The D's history was crossed against the opposing O's history and vice-versa. If at the conclusion of the game, the D had beaten their YTD avg, they went 'up a notch' in a power ranking. If the O did, then they went up a notch.
T-O margin I have to believe is a factor. Granted, most T-O's seem to be INT's. But those tend to occur most frequently when you're behind and the QB is forcing things. F-L kill a team in the redzone, so even though they occur less often, the impact is no less severe.
Lastly..FG % beyond 40 yds. If a team has a guy with good odds of kicking the long ones, they'll put some points on the board. Since points mean everything...this is IMO an often overlooked area.
|
| 13 | KrazyKoalaBears
Donor
ID: 266182910
Mon, Mar 25, 2002, 18:39
|
Sludge, thanks for the offer, but I'm not really concerned about predicting W/L. I'm just wondering what everyone has to say about what the most important factors are when they say a team is "great". For instance, if a team is 3-6, but all their losses were by 3 points or less, how to factor that in compared to a team that is 3-6 and has losses by 20 points or more. Obviously point differential comes up for this example and this is probably the easiest example. So I'm just wondering if there are other things like Yards, Time of Possession, and other things brought up, that you could also compare 2 teams with a similar record and determine which is "better" because this will lead to stats that can help determine "greatness". Again, thanks for the offer though.
|
| 14 | Toral
Sustainer
ID: 2111201313
Mon, Mar 25, 2002, 18:43
|
The 3 factors to use, then would be point differential, average-yards-per-play-differential, and turnover differential. As far as measuring superiority goes, gross yardage, TOP, first downs and the rest mentioned are, comparatively, garbage stats.
Toral
|
| 15 | Sludge
Sustainer
ID: 113368
Mon, Mar 25, 2002, 20:24
|
KKB - You asked what makes a team great. Finding out what best predicts whether a team wins or loses defines (in a sense) what makes a great team. If a team performs well in those stats which best predicts winning or losing, but the team still loses, then any scoring system based only on those stats will reward that team as if they had actually won (or will reward them nearly as well).
|
| Rate this thread: | If you wish, you may rate this thread on scale of 1-5. Ratings should indicate how valuable or interesting you believe this thread would be to other users of this forum. A '5' means that this thread is a 'must read'. A '1' means that this is a complete waste of time. If you have previously rated this thread, rating it again will delete your previous rating. If you do not want to rate this thread, but want to see how others have rated it, then click the button without entering a rating, or else click here. |
|
|
Post a reply to this message: (But first, how about checking out this sponsor?)
|
|