Which tie breaker is better, H2H or total points? (1 Viewer)

[spreagle]

I looked at our cumulative league historical records over several years (total points scored and combined wins and losses) and teams that scored 5% more than average had an overall 55% winning percentage. Scoring 10% more than average bumped that up to 60% winning percentage. It was the same thing on the downside. Scoring 5% less than average resulted in a 45% winning percentage and 10% less resulted a 40% winning percentage. For scoring, all teams were within 10% of average.

But of course year to year teams can have epic (or dreadful) scoring years. Scoring 15% more than average is a historical outlier and scoring 15% more usually only resulted in 65% winning percentages. This probably explains why teams can be the highest or second highest scoring in the league and still not make the playoffs. The chance of winning doesn't skyrocket for the highest scorers.

I don't know why scoring versus winning is so linear. It seems like at some point winning percentage should go up in a hocky stick fashion but it doesn't due to luck and week to week scoring variability. Anyways it seems to me tiebreakers should be on scoring because you can have an epic scoring years and that has so little of an impact on winning percentage (which I assume is head to head winning percentage also).

 

[Arodin]

I don't know why scoring versus winning is so linear. It seems like at some point winning percentage should go up in a hocky stick fashion but it doesn't due to luck. Anyways it seems to me tiebreakers should be on scoring because you can have an epic scoring years and that has so little of an impact on winning percentage (which I assume is head to head also).

Cummulative vs. real-time stats.

Think of the Arizona Cardinals last year. Great overall statistics, but hit the playoffs with some clown named Lindley (sp? not worth remembering) at QB and played like an 0-12 fantasy squad.

Team can run the fantasy regular season table, lose a couple key guys around week 12, and be a first-round out in the fantasy playoffs.

 

[rjv]

This should not even be a question.

Exactly

 

[spreagle]

I don't know why scoring versus winning is so linear. It seems like at some point winning percentage should go up in a hocky stick fashion but it doesn't due to luck. Anyways it seems to me tiebreakers should be on scoring because you can have an epic scoring years and that has so little of an impact on winning percentage (which I assume is head to head also).

Cummulative vs. real-time stats.Think of the Arizona Cardinals last year. Great overall statistics, but hit the playoffs with some clown named Lindley (sp? not worth remembering) at QB and played like an 0-12 fantasy squad.

Team can run the fantasy regular season table, lose a couple key guys around week 12, and be a first-round out in the fantasy playoffs.

I don't think key guys are as important in fantasy as in the NFL. The loss of a key guy can sink an nfl team but it takes the loss of several to sink a fantasy team, and a fantasy team always has a chance. What was eye opening to me when I looked at our numbers was that a fantasy stud like Gronk only adds +5% to the week to week fantasy chance of winning. I could draft an average team then steal Gronk from someones roster and only expect an 8-6 finish on average and maybe not make the fantasy playoffs (usually 8-6 teams have a 50/50 chance of making the playoffs in our league). I could steal three equivalent Gronks and only expect a 10-6 record and with bad luck and variability still not make the playoffs. Basically I need to be three Gronks or better than the league average at the start of the season to solidify a playoff chance. Or maybe anoer way of looking at it is to not lose a key player to injury while all other teams lose a key player, that in itself gives a two Gronk advantage I suppose.

 

[spreagle]

I changed my mind a little on on this. HTH could be a good option for dynasty/keeper leagues as it gives the perenially lower scoring teams a better chance to make the playoffs now and then. For league unity reasons, HTH may be a better option there. Redraft tiebreaker should be total scoring to reward the better teams.

 

[Adam Harstad]

I don't know why scoring versus winning is so linear. It seems like at some point winning percentage should go up in a hocky stick fashion but it doesn't due to luck. Anyways it seems to me tiebreakers should be on scoring because you can have an epic scoring years and that has so little of an impact on winning percentage (which I assume is head to head also).

Cummulative vs. real-time stats.Think of the Arizona Cardinals last year. Great overall statistics, but hit the playoffs with some clown named Lindley (sp? not worth remembering) at QB and played like an 0-12 fantasy squad.

Team can run the fantasy regular season table, lose a couple key guys around week 12, and be a first-round out in the fantasy playoffs.

I don't think key guys are as important in fantasy as in the NFL. The loss of a key guy can sink an nfl team but it takes the loss of several to sink a fantasy team, and a fantasy team always has a chance. What was eye opening to me when I looked at our numbers was that a fantasy stud like Gronk only adds +5% to the week to week fantasy chance of winning. I could draft an average team then steal Gronk from someones roster and only expect an 8-6 finish on average and maybe not make the fantasy playoffs (usually 8-6 teams have a 50/50 chance of making the playoffs in our league). I could steal three equivalent Gronks and only expect a 10-6 record and with bad luck and variability still not make the playoffs. Basically I need to be three Gronks or better than the league average at the start of the season to solidify a playoff chance. Or maybe anoer way of looking at it is to not lose a key player to injury while all other teams lose a key player, that in itself gives a two Gronk advantage I suppose.

I've looked at this before, too. So have several other smart guys, (wdcrob and ZWK, that I know of). Our findings are all wholly consistent with that. The best players over the course of a season are going to add 5-10% to your chances of winning a championship, (assuming you could acquire them at no cost whatsoever). People dramatically overrate the impact that a single player makes.

The best players during the fantasy playoffs will have a much bigger impact on your fantasy chances, (think: Odell Beckham last year), but people overrate how predictable those guys are. I remember in 2011 in one of my leagues, a team was starting Julio Jones and Darrius Heyward-Bey every week. During the regular season, they were both outside of the top 40 fantasy WRs; the team finished last in total points and snuck into the playoffs on tiebreakers. Then in the playoffs, Julio was the #1 fantasy WR and DHB was a top-10 WR, the owner had the high score in every week, and he walked away with the title.

Fantasy is weird. We always think it's going to be a lot more predictable than it really is.

 

[Deamon]

I prefer rock, paper, scissors for my league.

 

[Insein]

We do roshambo. I'm thinking of voting to change that though.

 

[Yenrub]

Of course people use head to head. It's more fun for more people when there are higher consequences for division rivalries and more strategy. Total points is better for people who are just gambling with anonymous strangers, but it's not really fun.

Your argument continues to fall down when the very real situation of the H2H game that decides the playoff spot is played in week 5 is considered. How is it more "fun" to decide the playoff team two and a half months later based on who won that game? Your argument that you somehow manage your team differently tomaccount for division games is also weak. What exactly do you do differently in your week 2 division game than you otherwise would?

The fact is that total points identifies the better team. Fantasy football has a very large element of luck to begin with. To the extent that elements of that can be eliminated to identify the best players, that should be done imo.

I have no problem either way, to each his own

I disagree that total points identifies the better team though

A team that had Big Ben as their top QB and had to back Ben up with say Alex Smith is probably going to be outscored for the season by a team that started Eli at QB all season long.

All other positions being equal I think the team with a healthy Ben is a “better” fantasy team.

 

[Don Quixote]

All-play record.

 

[xenith]

i like to combo approach, H2H and an all play type hi score hybrid.

I do have to say though, you guys/gals in here whining about H2H when too many players are on BYE or just an unlucky loss, YOU are the ones crying for the participation trophies. YOU LOST!!! "But i don't care if i lost, I'M A WINNER DAMNIT!!"