But still needs work.
The problem is this:
Stats is basing the lifetime rankings on 'average batting points' plus 'average hitting points' by combining the number of batting points scored last year plus this year divided by 2. (And pitching+pitching divided by 2)
The problem is that last year there were 195 teams and this year there are 300 teams in the competition. Taking my average from last year in every category on a scale of 1-195 and averaging it to this year on a scale of 1-300 is not accurate. Stats needs to make the seasons equal or as the contest grows in the future all the earlier years will be discounted.
Last year scoring a 150 in HR (out of 195) was very solid work. This year scoring a 150 (out of 300) is very average!
Thanks,
Dyv
Overall Lifetime Ranking Better....
Overall Lifetime Ranking Better....
Just Some Guy
Overall Lifetime Ranking Better....
STATS has got it all wrong. Lifetime standings should be based on overall finish, not batting average, batting ranking, ERA, pitching points, or who's father can beat up the other's fathers.
It's all very easy, not complicated, and fair. This is how it should be done.
Last season, if you finished first, you get 1 point. If you finished tenth, you get 10 points. If you finished 195th, you get 195 points.
This season, if you finish first, you get 1 point. If you finish tenth, you get 10 points. If you finish 195th, you get 195 points, and if you finish 300th, you get 300 points.
The only "lifetime" standings participants should be those that have played both seasons.
Add the final standings points up, and the lowest number is the leader. For purposes of mid-season standings, continue to use last years numbers, and use the current numbers for this season.
Without comparing it to the BCS poll, it is the same idea. It doesn't matter how many votes a team gets in a poll, all that matters is the teams' overall ranking in the poll.
The same should be used here.
Oh, by the way, the original post for a lifetime standings was mine, and I never invisioned that STATS would have basterdised it as they have.
However, with my team sitting in 196th this season (after finishing 12th last year), I am beginning to think that the idea of a lifetime standings isn't so good anymore.
Buster
It's all very easy, not complicated, and fair. This is how it should be done.
Last season, if you finished first, you get 1 point. If you finished tenth, you get 10 points. If you finished 195th, you get 195 points.
This season, if you finish first, you get 1 point. If you finish tenth, you get 10 points. If you finish 195th, you get 195 points, and if you finish 300th, you get 300 points.
The only "lifetime" standings participants should be those that have played both seasons.
Add the final standings points up, and the lowest number is the leader. For purposes of mid-season standings, continue to use last years numbers, and use the current numbers for this season.
Without comparing it to the BCS poll, it is the same idea. It doesn't matter how many votes a team gets in a poll, all that matters is the teams' overall ranking in the poll.
The same should be used here.
Oh, by the way, the original post for a lifetime standings was mine, and I never invisioned that STATS would have basterdised it as they have.
However, with my team sitting in 196th this season (after finishing 12th last year), I am beginning to think that the idea of a lifetime standings isn't so good anymore.
Buster
-
nydownunder
- Posts: 522
- Joined: Tue Jan 25, 2005 6:00 pm
- Contact:
Overall Lifetime Ranking Better....
Buster
What happens at the completion of year 3 and someone who hadn't particpated in year 1 has finished 10th and 20th respectively. Is this person not considered one of NFBC all time best players? And besides, your pool of lifetimers will decrease ever year...so what's it really telling you when your #50 after year 3 when there are only 60 left in your sample. This would also mean that you could move up year to year purely based on the fact that someone has dropped out of the competition. In my industry we call this 'Survivor Bias' and is considered not to be an accurate illustration of the facts.
I think the current format, with Dyv's suggestions keeps everyone interested. And you never know, Greg and Tom may come up with an interesting way to reward the leader at season's end...assuming a mininum of year's requirement (ie >1) to avoid 1st year or 1 year players winning the award. For example, end of year 3 all two year players are eligible for award: assuming they are in that year's comp. And yes, I would argue someone having 2 very good years and not having played a 3rd year (ie year 1), is deserving of such an award.
[ June 09, 2005, 01:04 PM: Message edited by: nydownunder ]
What happens at the completion of year 3 and someone who hadn't particpated in year 1 has finished 10th and 20th respectively. Is this person not considered one of NFBC all time best players? And besides, your pool of lifetimers will decrease ever year...so what's it really telling you when your #50 after year 3 when there are only 60 left in your sample. This would also mean that you could move up year to year purely based on the fact that someone has dropped out of the competition. In my industry we call this 'Survivor Bias' and is considered not to be an accurate illustration of the facts.
I think the current format, with Dyv's suggestions keeps everyone interested. And you never know, Greg and Tom may come up with an interesting way to reward the leader at season's end...assuming a mininum of year's requirement (ie >1) to avoid 1st year or 1 year players winning the award. For example, end of year 3 all two year players are eligible for award: assuming they are in that year's comp. And yes, I would argue someone having 2 very good years and not having played a 3rd year (ie year 1), is deserving of such an award.
[ June 09, 2005, 01:04 PM: Message edited by: nydownunder ]
Wagga Wagga Dingoes (NY#4)
Luck is where preparation meets opportunity!
Luck is where preparation meets opportunity!
Overall Lifetime Ranking Better....
Actually, it still works. The option is to do an average, or take the best of the two or three or four, etc. years.
You have to reward those who play more than one season. So, after two years, you take either the average of the scores, prorated, or drop the lowest score.
Thus, if you miss year one, but play in years two and three, and finish 10th in year 2, and 20th in year 3, you have an average of 15.
If someone plays in all three, finishes 10th in year one, 20th in year 2, and 30th in year three, you have either a) an average of 20, or if you drop the low year, an average of 15.
You can't base lifetime achievements on less than a lifetime, and you have to award those that continue to come back. If the competition continues for a number of years, you limit the lifetime awards to those who have played, say, five years.
You can still average whatever you want, but you need to do so based upon final standings, not how many points you earn in hitting, pitching, or fishing.
You have to reward those who play more than one season. So, after two years, you take either the average of the scores, prorated, or drop the lowest score.
Thus, if you miss year one, but play in years two and three, and finish 10th in year 2, and 20th in year 3, you have an average of 15.
If someone plays in all three, finishes 10th in year one, 20th in year 2, and 30th in year three, you have either a) an average of 20, or if you drop the low year, an average of 15.
You can't base lifetime achievements on less than a lifetime, and you have to award those that continue to come back. If the competition continues for a number of years, you limit the lifetime awards to those who have played, say, five years.
You can still average whatever you want, but you need to do so based upon final standings, not how many points you earn in hitting, pitching, or fishing.
-
nydownunder
- Posts: 522
- Joined: Tue Jan 25, 2005 6:00 pm
- Contact:
Overall Lifetime Ranking Better....
I agree with your last point, as I personally would not care how I won (by category), just that I won.
Wagga Wagga Dingoes (NY#4)
Luck is where preparation meets opportunity!
Luck is where preparation meets opportunity!
Overall Lifetime Ranking Better....
Originally posted by Buster:
Actually, it still works. The option is to do an average, or take the best of the two or three or four, etc. years.
You have to reward those who play more than one season. So, after two years, you take either the average of the scores, prorated, or drop the lowest score.
Thus, if you miss year one, but play in years two and three, and finish 10th in year 2, and 20th in year 3, you have an average of 15.
If someone plays in all three, finishes 10th in year one, 20th in year 2, and 30th in year three, you have either a) an average of 20, or if you drop the low year, an average of 15.
You can't base lifetime achievements on less than a lifetime, and you have to award those that continue to come back. If the competition continues for a number of years, you limit the lifetime awards to those who have played, say, five years.
You can still average whatever you want, but you need to do so based upon final standings, not how many points you earn in hitting, pitching, or fishing. You do have to be careful because you run the risk of the same person winning the lifetime award every year, lol.
My suggestion would be to have a rolling 3 year average.
Either way - the ranking IS the indicator, but I can live with the points scored method. After all, points scored is what determines ranking. Overall ranking doesn't care if you had pitching or hitting or fishing... points are points.
It just needs to reflect something real and accurate and I'm content that it has some value. Right now it's ridiculous
Dyv
Actually, it still works. The option is to do an average, or take the best of the two or three or four, etc. years.
You have to reward those who play more than one season. So, after two years, you take either the average of the scores, prorated, or drop the lowest score.
Thus, if you miss year one, but play in years two and three, and finish 10th in year 2, and 20th in year 3, you have an average of 15.
If someone plays in all three, finishes 10th in year one, 20th in year 2, and 30th in year three, you have either a) an average of 20, or if you drop the low year, an average of 15.
You can't base lifetime achievements on less than a lifetime, and you have to award those that continue to come back. If the competition continues for a number of years, you limit the lifetime awards to those who have played, say, five years.
You can still average whatever you want, but you need to do so based upon final standings, not how many points you earn in hitting, pitching, or fishing. You do have to be careful because you run the risk of the same person winning the lifetime award every year, lol.
My suggestion would be to have a rolling 3 year average.
Either way - the ranking IS the indicator, but I can live with the points scored method. After all, points scored is what determines ranking. Overall ranking doesn't care if you had pitching or hitting or fishing... points are points.
It just needs to reflect something real and accurate and I'm content that it has some value. Right now it's ridiculous
Dyv
Just Some Guy
Overall Lifetime Ranking Better....
The way it should be done.... haven't had the time to read the threads on this, so if I repeat something, get over it.
Each year of rankings needs to be pro-rated based off the highest # of entries in whatever year that occurs.
Right now, we have 195 and 300.
If next year, it's 400, then everything 2004 and 2005 prorated to 400 teams.
Here ya go.
2005 - Simply give #1 team 300 "rating" points.
Give #300 1 "rating" point.
2004 - Give #1 "300" rating points (Hi so far)
Give #2 "298.462" points.
Give #195 "1.538" points.
300/195 = 1.53846 (Every 2005 pt is worth 1.53846 2004 pt)
Add up the points each year!!
Re-"pro rate" 2004 and 2005 next year based on # of entries next year if over 300.
This rewards those that play every year.... unless they really suck!! It could reward a guy in a few years that missed 2004, but was something like Top 10 2005-2007. If a guy doesn't play, he gets zero.
Never throw any low/hi out. That's bs.
Each year of rankings needs to be pro-rated based off the highest # of entries in whatever year that occurs.
Right now, we have 195 and 300.
If next year, it's 400, then everything 2004 and 2005 prorated to 400 teams.
Here ya go.
2005 - Simply give #1 team 300 "rating" points.
Give #300 1 "rating" point.
2004 - Give #1 "300" rating points (Hi so far)
Give #2 "298.462" points.
Give #195 "1.538" points.
300/195 = 1.53846 (Every 2005 pt is worth 1.53846 2004 pt)
Add up the points each year!!
Re-"pro rate" 2004 and 2005 next year based on # of entries next year if over 300.
This rewards those that play every year.... unless they really suck!! It could reward a guy in a few years that missed 2004, but was something like Top 10 2005-2007. If a guy doesn't play, he gets zero.
Never throw any low/hi out. That's bs.
Overall Lifetime Ranking Better....
Originally posted by UFS:
The way it should be done.... haven't had the time to read the threads on this, so if I repeat something, get over it.
Each year of rankings needs to be pro-rated based off the highest # of entries in whatever year that occurs.
Right now, we have 195 and 300.
If next year, it's 400, then everything 2004 and 2005 prorated to 400 teams.
Here ya go.
2005 - Simply give #1 team 300 "rating" points.
Give #300 1 "rating" point.
2004 - Give #1 "300" rating points (Hi so far)
Give #2 "298.462" points.
Give #195 "1.538" points.
300/195 = 1.53846 (Every 2005 pt is worth 1.53846 2004 pt)
Add up the points each year!!
Re-"pro rate" 2004 and 2005 next year based on # of entries next year if over 300.
This rewards those that play every year.... unless they really suck!! It could reward a guy in a few years that missed 2004, but was something like Top 10 2005-2007. If a guy doesn't play, he gets zero.
Never throw any low/hi out. That's bs. Simple but good, I like it.
The way it should be done.... haven't had the time to read the threads on this, so if I repeat something, get over it.
Each year of rankings needs to be pro-rated based off the highest # of entries in whatever year that occurs.
Right now, we have 195 and 300.
If next year, it's 400, then everything 2004 and 2005 prorated to 400 teams.
Here ya go.
2005 - Simply give #1 team 300 "rating" points.
Give #300 1 "rating" point.
2004 - Give #1 "300" rating points (Hi so far)
Give #2 "298.462" points.
Give #195 "1.538" points.
300/195 = 1.53846 (Every 2005 pt is worth 1.53846 2004 pt)
Add up the points each year!!
Re-"pro rate" 2004 and 2005 next year based on # of entries next year if over 300.
This rewards those that play every year.... unless they really suck!! It could reward a guy in a few years that missed 2004, but was something like Top 10 2005-2007. If a guy doesn't play, he gets zero.
Never throw any low/hi out. That's bs. Simple but good, I like it.