I am starting work on an article that I want to write for this site, and I need some help. Daullaz, who is a member here, developed an excellent formula for quantifying the degree to which luck sways standings in H2H, point leagues. I want to take his formula and apply it to data to a sample large enough to allow me to make inferences about the entire population of H2H, point leagues.
I currently have data from 5 leagues. I need data from at least 25 more before I can start making any reasonable assertions about the population. More than that would be excellent, though.
If you played in a H2H, points league last year and want to help me out, here's what I will need from you.
The amount of points each team in the league scored in each week of the regular season.
The final, regular-season standings of the league.
I know that is a lot of data to ask each of you to post. If the data is available on a website somewhere, a simple link will suffice. The league can be any size, and the regular season can last any number of weeks. What matters is that the data comes from a league with a H2H format and points scoring.
Doing one for football, and can say it stinks big time.
Yes doctor, I am sick. Sick of those who are spineless. Sick of those who feel self-entitled. Sick of those who are hypocrites. Yes doctor, an army is forming. Yes doctor, there will be a war. Yes doctor, there will be blood.....
I don't know if simple points will tell you about luck. It will give you a good start but, not the whole story. i am trying to find my points league stats for you. this year I finished 6th in points in an 18 team league. I finished dead last in record, 8 games out of first. During the season i lost 8 games by 10 points or less and 5 by less than 5. Most frustrating loses. Losing by one when Cameron breaks face. Losing by tie breaker when corey patterson gets demoted. (Lineups lock for the week on Monday mornings). Lose by 3 when Prior gets drilled by line drive. These are tuff loses that will show up in the numbers, but especially in my league there are times where players are only expected to miss a couple of days on Monday morning, but don't end up playing all week. this also happened to me a lot this year. chipper and Thome killed me with this. There were a couple of weeks were I lost by 30 -40 playing 1-2 guys down. A decent week by an offensive player in the league is 25 - 35. I also caught Jon garland and Dontrelle Willis when they put up 100 point weeks.
I am very interested to read that article, and see that formula. My guess is luck is 75% of a head to head points league.
rob0417 wrote:I don't know if simple points will tell you about luck. It will give you a good start but, not the whole story.
Yeah, luck actually isn't the best term to use. I just used it because so many people associate luck with H2H, so I figured it would get more people's attention. Literally, the calculation determines the effects that the schedule has on your overall record. However, you have zero control over your schedule, so - to some degree - it is a measure of luck.
rob0417 wrote:i am trying to find my points league stats for you.
rob0417 wrote:I am very interested to read that article, and see that formula. My guess is luck is 75% of a head to head points league.
Early numbers are leaning towards an average impact of about 10% one way or the other. That means that in a 20-week season, your record could be swayed by as much as 2 wins or losses because of the schedule.
Again, that's the very early result of a very small sample. The impact could drastically increase or decrease as I incorporate more data. The final numbers will be part of a bigger article I am going to do in an effort to simply get some fundamental information out to the masses regarding H2H, points leagues.
StlSluggers wrote:Again, that's the very early result of a very small sample. The impact could drastically increase or decrease as I incorporate more data. The final numbers will be part of a bigger article I am going to do in an effort to simply get some fundamental information out to the masses regarding H2H, points leagues.
I understand that you are highly defensive of H2H points league, but the early returns makes me very skeptical. I'm not suggesting that you are creating some grand subterfuge, but I think its going to be very tough to accurately define (even somewhat closely) what you laid out. Or maybe fantasy football H2H points is so ridiculously different that it doesn't deserve to draw any sort of comparision. I've only played one H2H points baseball league and it was nothing more than filling in for a friend who couldn't do the league any more and was well out of contention when I took over. I honestly didn't pay a ton of attention to it so I'm not professing to be a master of the game design.
My quick formula on the value of scheduling in H2H (this is based on Fantasy Football through 10 weeks with 12 teams or 6 games per week)
Number of times a team lead the league in scoring for that week and consequently would have won no matter the matchup: 10
Number of times that a team scored the lowest number of points in the league and would have lost no matter the matchup: 10
20 (or 33%) of the games are decided without regard to the schedule. The other 40 (or 66%) are completely based on the schedule. Times that the highest scoring team faced the lowest scoring team would change the formula some, and the number of teams would definitely change the formula (no matter how many teams there are, only one can score the most/fewest points).