Musings on ROI and the economics of staking arrangements

[nsidestrate]

This is not intended as a criticism of any specific staking arrangement or any player on the forum. I tried to pick a time to post it when there were no active fundraising efforts, so it wasn't seen as a slam on any particular player. I have just been thinking about the math of staking in the abstract and wondered what a realistic EV estimate is for various staking deals.

I'm going to start with the assumption that a very good player of the sort who is likely to seek staking has a true ROI of 30%. I believe that there might be some players whose long term ROI is more like 70%, but I think that is close to an upper bound for even "superstar" players over millions of events. I've seen arguments that a superstar might be able to hit 80% ROI, but looking at OPR for players with tons of MTTs I can not find good evidence to support that view. Nobody really has enough data, but superstars like Rizen or SCTrojans run at 67% or 71% ROI. Guys who play even more, like PearlJammer or BelowAbove run 49% ROI. I know a lot of players that I believe to be quite good who have lifetime ROI figures that are in the teens.

I've seen a number of staking deals where the backer puts up 100% of the money and gets 60% or 50% of the payout. In the case of our hypothetical backee, a $100 investment should result in $130 being earned. If the arrangement doesn't include 100% return of buy-ins before the split, you expect to lose 22% of your money on every dollar invested at 60%. Obviously, an arrangement that doesn't pay back the buy-ins first is a terrible deal unless the player has a stratospheric ROI. He would need an ROI of 67% to break even for the backer if 60% is paid out. If the arrangement is 60% of the profit (all buy-ins paid back first), a 30% ROI becomes 18% to the backer -- 50% would obviously return an ROI of 15%.

Even if all the money is returned to the backer until the buy-ins are paid back, the EV of the deal is probably less than the backer thinks. The reason has to do with the nature of wins and losses in MTT play. Let's assume we have a MTT player whose ROI is 30% and he plays 10 sets of 50 $11 180s. Each set of 50 180s would cost $550 to his backers. The run of 10 180s will cost a total of $5,500 in buy-ins and his ROI says that he should average $1,650 in winnings over that span. Of course, 500 MTTs is nothing compared to the long run, so his actual results could easily be up or down, but we'll pretend that he wins exactly his ROI, so his net profit will be $1,650 over the series. This implies a $990 payout to his backers (if it was all accounted for in one lump, it would be the case). However, if each set of 50 180s is accounted for on its own, a bad trend will begin to emerge. Let's imagine a hypothetical string of results:

-$100
-$40
+$1,200
-$160
-$80
-$105
-$140
+$360
-$85
+$800

In this specific case, instead of the $990 that the backer would expect for 60% of the net winnings, the backer would only clear $706 (or 43% of the winnings or an ROI of 12.8%). If the deal called for an equal split of profits, the backer would clear $470 (or 28% of the winnings or an ROI of 8.5%). This phenomenon is due to the fact that the backers absorb 100% of the losing sessions, while only booking 60% of the winning sessions. For this reason, backers in $10,000 events usually arrange for backing fees for past events to be pulled from future winnings.

The situation gets worse if your stakee only has an ROI of 10%. Using the same 60% to the backer, 40% to the player division and assuming a string of results like this:

-$100
-$40
+$900
-$160
-$80
-$105
-$140
-$200
-$85
+$560

You would see a net profit of $550 for an overall ROI of 10%. The player would actually pocket $584 ($360 from the +$900 set and $224 from the +$560 set). The backer would lose $34. Now these numbers may not be realistic, but they illustrate a point about the dangers of backing arrangements. In fact, the idea of backing a series of 50 180s is to try to smooth out the variance and to get a more consistent ROI across the board for each set. Obviously, we don't have nearly enough data to determine if this is actually possible. I can tell you that even a player like Rizen, who I assume we would all agree is an excellent MTT player has an ROI on Stars of -31% over the past 120 days which includes 329 events.

I'm not trying to stop backing or even really to discourage it, I just want people to understand that they are primarily engaged in community building and not bankroll building.

[MXRider]

Firstly, I think that we can all agree that the staking up to this point has been, for the most part, all about a sense of community and comraderie. I know it has been from my standpoint in both the stakes I have received and those that I have participated in as well. I think it goes even further to show how much of that exists in the stakes that have been arranged and those that have failed.

Now, with that said, I think your comments are quite true, but somewhat limited as the size of the field and the opponents we face. I truly believe, someone could honestly get to a 120% ROI in the 4/180s but have that fall off sharply at the 20's. A slight change in skill level within a tournament can drastically impact the anticipated ROI over the long term.

Not sure if that quite makes sense or not, but I think it does.

[mclelands]

I have nothing to add except impeccable timing, Nside

[nsidestrate]

[MXRider]

[kinnipak]

As a relative newbie to ITH, I might not be the best qualified to answer this (not sure it even requires an answer) but I would like to make a comment as I have staked a few here on ITH.

My reason for staking is purely and simply to become a better poker player and be able to contribute to ITH in some small way. Staking enables me to watch others play when I can't, and learn from the way play goes. I have defiantely become a better player for it. If some money can be made, well, who would be against that? But I sincerely hope that staking is done with the right intentions and knowledge of the pitfalls that can occur. I, for one, would never question the play or motives of someone I have staked. I root for them when they run hot or cold.

I read nsides's post as a feeler to make sure that one knows what one is getting into before just giving someone money. Which is certainly something that everyone should know before they consider staking someone. But in my newbie opinion, staking is definately a +ev thing for a solid poker community.

Just my 2 cents.

and NO, I do not know how to spell "definately".

[mclelands]

[PauliF]

you ever seen the producers?

from next year i plan a similiar scam
i will get well known around a variety of online and live poker communities
then i will get staked in major tournaments
i can talk the talk
i know my Ms from my patience factors
i can do simultaneous equations
i know pretty much all the odds and that you need to know
so can talk the talk
then i would play the way i have been playing these last two years
propper gamblin fever tastic
so i would get myself staked to maybe 300-400%
never money and have a great free holiday

actually it started off as a joke but i might not post this now
quite a good idea

[MXRider]

[toom]

I've been thinking about making a post like this since we did the first Million stake last year. The founding of the DAQ made me think about it more. I had always intended to make such a post, and never got around to it. Thanks for starting the discussion.

Since MTTs have such variance, and good players only "win" (ITM) 20 or so per cent, that makes it all about the long term. (Duh.) That makes staking a series (like the 50 4/180 stakes) more likely to be profitable than staking a single tournament.

This thread illustrates the value of the SharkDAQ.

If you're staking a single tournament, you should expect to lose. If you stake a series, you should expect to (approximately) break even. But staking several series, you will roughly break even, while waiting for the one big win that makes it all worthwhile.

[taz115]

This analysis should only be applicable to arrangements where backers are putting up 100% of the buy-in. When the buy-in is split between backers and stakers and the profits are paid off 1:1 everyone should have the same long-term EV.

If the players asks for a premium, his EV is increased at the expense of the backers.

[rocketplayer]

[kinnipak]

[torch]