For those that have been following along with the blog so far this year, you'll know that I mostly join head to head contests as my choice of contests for cash games. I have explained it a little bit in a few different posts but have not dedicated an entire post to it yet this season. The primary reason that I join head to head contests is to mitigate the risk caused by 50/50's and double ups. In 50/50's and double ups no matter how many you join it is an all or nothing strategy each week.
There is nothing worse than finishing in the 49th percentile of lineups in a 50/50 or Double Up and losing almost all of your money invested in a given week. Having this happen a few weeks in a row back in 2014, where I just missed the cut line in my 50/50's, I decided to make a change in the way that I am playing. I am able to get more of my bankroll in play each week if I join a high number of H2H games as the overall risk to my bankroll is lower since it is not an all or nothing strategy. Ultimately the downside of the decreased risk is the decreased upside, if I have a lineup that is in the 80th percentile, I am not going to win all of my contests, I am likely only going to win 80% of my contests, where a 50/50 I would win all of my contests. Ultimately, I am ok with the decreased upside for the decreased risk, as the goal is to grow the bankroll and not let a few bad weeks kill your season.
A key example of this is week 2, in week 2 my lineup finished in about the 40th percentile in 50/50's and double ups consequently, I still won about 40% of my contests which had I joined only 50/50's I would have lost pretty much 100% of my investment.
This is interesting Devin. I just can't get into the H2H. Last week I won every 50/50 and DU yet only half of my H2H. It was the same deal last season. Am I doing something wrong?
ReplyDeleteCouple of things, first, what place did you finish in that 50/50? I would guess you finished towards that middle range which I referred to in the post. Second, how many h2h are you playing? To get a proper sample size you have to play a significant amount of h2h games otherwise I recommend playing 50/50's. On a given week I'll have 50+ h2h games to get that true variance of opponents to give me a representative sample equivalent of a 50/50.
DeleteThanks for your response! I only played two lineups last week and entered them in 10 H2H each (20 total, all $1 and $2). Maybe I'm not playing enough. Lineup 1 went 6-4 H2H and finished 10th to 25th in 100 player 50/50s. Lineup 2 went 5-5 in H2H and finished 24th to 43rd in 100 player 50/50s. Interestingly, the latter lineup finished 17th to 30th in 100 player top 12s.
DeleteInteresting, without knowing the exact scenario, I would say you either got unlucky with your head to head matchups statistically. Either way the head to head strategy is just mine finding a strategy that works for you is ideal as there is not one that is significantly better than another. Mine is more risk adverse, as think of it is as if you finished 55/100 in a 50/50 but still went 4-6 in H2H, so you would have lost all of your 50/50's but still won 4 of your h2h.
DeleteMy biggest issue with H2hs is that a great lineup can lose. I used to play tons of H2Hs but slowed down on them for that reason and now concentrate the majority of my entries on 50/50s. Nothing worse IMO than having a great line lose because some fluke player has the game of his life. I guess I can see both sides of the coin
DeleteI've been playing Thursday lock contests on FD now that you can late swap and I've noticed the cashlines to be lower for Thursday contest both weeks. Week 01 was a bit strange since NE players were super highly owned and almost all busted, but even Week 02 was lower (only notably play was CIN DST at 10% ownership).
ReplyDeleteFor reference, I played the same lineup in Main slate 50/50s and Thurs-Mon slate 50/50s and score 127 -- which was ~26th percentile for main, but ~20th percentile for Thur-Mon.
I think a lot of the super chalky plays aren't really known until later in the week by most people and I can imagine that some people never go back and update their lineups for Thursday lock. Seems like an edge to me, but will collect a few more weeks data before going all-in.
It really depends on the Thursday game. For example week one, the Patriots drew heavy ownership in week one and then in week 2 people will always play the Thursday or Monday night games both were clunkers this week.
DeleteA few sample ownerships:
ReplyDeleteJulio Thurs -> 50-60%
Julio Main -> 60-70%
TyMont Thurs -> 50-60%
TyMont Main -> 65-75%
Thurs cashlines averaging about 109-111
Main cashlines averaging about 116-118
It really depends on the Thursday game. For example week one, the Patriots drew heavy ownership in week one and then in week 2 people will always play the Thursday or Monday night games both were clunkers this week.
DeleteHello,
ReplyDeleteThank you for your advice. It has been very helpful. When you play multiple H2H's do you primarily use the same line-up in each contest or do you use variations of that the line-up? Thanks again.
Hey David,
ReplyDeleteI trying to figure out where the benefit is coming from when choosing H2H's over 50/50's. Personally, I've found 100 team 50/50's to be the most advantageous over the long run, as the risk of losing is a lot less. From a conceptual way of thinking about it, I've always found that in a H2H, I am playing to beat another single owner, and if that one owner makes great picks, I'm toast. Whereas, in a 100 team 50/50, I'm more playing to not do worse that 50 other owners. If one of those owners makes a great pick, I still have 99 other chances of either them making bad picks, or their player getting injured, etc. There's also quite a bit of statistical analysis covering this debate. I'm by no means trying to prove you wrong. I'm just interested in learning if there is another side of the debate I'm missing. Great blog, though. Thanks for the advice.
It has everything to do with scale of how many h2h you're going to play. If you play enough h2h volume. Think of it this way and using an extreme example here. If you were to play a h2h against every person in that 50/50 that you mentioned and you finished 51/100 in the 50/50, you would still have won 49 of those h2h contests.
ReplyDeleteOk, I see where you're coming from now. When you take volume in to consideration, then I can see where it can make sense. I also see why I didn't get it to begin with ... I don't play that many h2h at all.
DeleteWhen I enter h2h I always got the impression Im going vs a shark. You look at their experiences and almost all of them have the all the checkmarks checked. Maybe Im wrong, but like the poster above I feel 5050s are by far the best chance at cashing
ReplyDeleteI am actually running an experiment to test this during this season. It is comparing the average h2h scores I have compared to the cash lines in 50/50's should be an interesting test and once I have more data, I will share.
ReplyDeleteNice experience!
Deleteyeah this is a big consideration. your assumption that youre better off playing h2hs assumes that all your opponents will be of equal strength. i think common sense dictates that this just isnt true...
Deletethe guys posting h2hs in the lobbies are, generally speaking, going to be better players than those that join 50/50s. they just have to be.
think about it. the 'fun players' if you will that you most want to play against are new players, guys who dont update their lineups, guys that dont understand all the value plays, etc. that just isnt the profile of a guy who posts a hu contest, even at the smallest buyin levels.
Yeah, I typically avoid the top end players to try to weed those guys out. I get your point, will be interesting to see how the experiment comes out regarding what truly is more beatable.
DeleteThis is an intuition I've been having too. Upon further thought, though, I'm not sure it's actionable.
DeleteWhat's the maximum difference in the average cash line in a large 50/50 vs. H2Hs, on average? Maybe 3-5 points? For me, if my lineup is even that close to the cash line, I would much prefer the safety of the H2H strategy vs. the all-or-nothing 50/50.
Great blog Devin, would love to hear your thoughts.
Devin do you create h2h's? or join existing h2h's? I ask because it seems my opponent has been more mediocre when i create and let them come to me..compared to joining vs someone who seems like a shark
ReplyDelete