Mathematics of Spin & Go tournaments on PokerStars

Welcome to the forums!

I'd think for cash game players there will be quite a hard transition period, especially if you haven't played a lot of headsup. The games are completely different when you're typically maximum of 25 big blinds deep and on average I think you're about 13-14bbs deep during the game. If you're a bum hunter type, spins are not in general for you, as there is now no possibility to game select. Problem for spins are, that even if you're having a certain ROI, if you're not hitting the highest multipliers, the ROI you're going to get is going to be 2% less. Generally I have said that these games are grindable if you're making 4% ROI or more, so that you can almost every month expect to make money. I'd think that the best players playing only a few tables are making somewhere in between of 6-8% in the lower stakes, but I'd think 4% is quite close to what is achievable at 30$ and up.

Right now it's still a bit in the air what is possible since PokerStars has removed all the options to game select and I'm following it with interest how regs ROIs will change now.

What kind of deep strategy is needed is just to be really good at short stacked situations. Format is closer to hyper HUSNGs, for me it was quite a journey to unlearn everything about ICM while multitabling these (and then when I got back to 6-max and 9-max hypers, it has been a journey again to relearn ICM...).

I'd think these games are really far away from just Nash, the reason mainly being that when you'll need to achieve 9-11% edge in just 20-25 hands, you'll need to be very exploitative. I'd think it's mostly an exercise to have a really solid standard strategy that exploits usual opponents and then adjust to being more unexploitable against better opponents.

You'll pick up population tendencies by analyzing them, practically it takes more time than in 6-max as you'll get only 1-2 opponents with you on the table at the same time. In general I would think that for you it would mean a lot of basics work on playing shortstacked.

As Ghaleon said, for a cash game players these games are from another planet with their variance. You can be 500 buyins under EV in 10000 games and you can be 300 buyins under EV in a single month. Just a friendly warning.

I'm still offering coaching for spins as well and can help you with the analysis part and shortstacked part a lot.

 

I'd think this is mostly a comparison of results lines, not EV lines. If you're looking at EV$ line of a profitable SNG player and a EV graph from chipev at Spins graph, I'd say that my Spins EV is steadier. Even the EV line with multipliers taken into account isn't that different from a SNG EV graph.

Results on the other hand, for sure are not steadier. Feelings are going with the same rollercoaster in Spins.

 

I think on our yesterday's session I stumbled upon a good and easy explanation why I don't think the million dollar SNG is a promotion at all. PokerStars charges 1% higher rake for million dollar Spins than they charge for comparable buyin level SNGs.

Every time a million dollar award is given out (3 out of 10 million in 5$s, 3 out of 100 million in 0.5$s). In 0.5s, it means 100 million extra rake percentages taken and in 5$s it means 10 million extra rake percentages taken. So every time PokerStars gives out a million with this promo, they have made 500000$ (half a million) extra.

For one million to drop, in 0.5$s it means that PokerStars has raked in 4 million for those games (and probably returning a very small percentage back to low VIP level players) and in 5$s it means that PokerStars has raked 3.5 million in the games (again, probably very low average VIP level, very low rewards). It's a damn good promotion for PokerStars. They're making a ton of money for themselves and people are thinking that it's a great promotion and PokerStars is giving out money. Very smartly executed.

 

Some of the calculations are a bit outdated, I should pretty much just start from scratch and update everything. I will try to do it in the near future, but I don't think I can do it during this year. A slight problem with a new format since the rules keep changing a lot. It will still be around where it should be, like % you need to be a winning player should be there. Also everything on 3000x Spins (like the bankroll management parts) applies to 10000x Spins as well.

I don't think it changes a lot for low stakes Spins. It will change a ton for the players that are now doing SNE with the highest stake spins though.

I would be worried if I would be maining Spin & Gos now, since they are pretty much the most opaque format available. It is really hard to see when the population changes so much that the games will be unbeatable for you and it requires a lot of volume to determine it. A lot of volume in a high variance game can be really costly if your edge has disappeared.

 

I think I have played a lot with both Koovoon and Play2win before at Spins, so at least a two reg game. Oh well, when there's 1 million as main prize, perhaps even 2 reg game can be good

 

For 100$, it's the same as in 15$, 30$ or 60$ games, so you'll need to win on average 26.5 chips every game. To make these games grindable, in general you would like to see at least 3-4% ROI.

For 5$, it's the same as in 1$ spins, so you'll have to win 38.5 chips every game. Practically to make these games grindable, in general you would like to see at least 5-6% ROI - in these games a lot more is given to the top, so it will increase the pressure you're having here.

For 0.5$ we don't have ready calculations yet. Practically you would need to finish 36.3% ITM to be breakeven and that would be about 43.5 chips per game. Same caveat applies as for 5$ games, in general you would like to see 5-6% ROI for these games.

 

I believe at 15s there are about 20 games running per minute. They can give the illusion of games popping up instantly by being a bit overly hopeful in opening the game and they are having plenty of delays in the system where they're starting to open the game so that it still feels instant. At least I remember situations when the table has opened for me, but there's just one player with me waiting for a third or I'm sitting a few seconds alone. Especially if starting a session by opening 5 tables instantly. Might be just 30s though.

So I believe there should be about 4 120x+ games running per day in 15s and about 1 240x running per day.

 

I'd think one good point would be to check tournament numbers. I used two hypers that start pretty close to when they are created. One of them was played today, an earlier one on 7.11, so there's 16 days in between. During those 16 days, there were 15.7 million games played on PokerStars.

So there are about 700 tournaments going on every minute. Is about 1 out of 3 of those games 15$ spin & go? I'd even believe that 5$ spins would be about 5 times more popular than 15$ games. And 0.5$ million dollar spins even more popular than 5$ ones, probably by a factor of 2 or 3.

Even having about 20 of the games going on every minute would mean every 35th game that is running on PokerStars would be a 15$ Spin & Go. I wouldn't be surprised if Spin & Gos would be the majority of running tournaments on Stars.

 

It's a line that takes away some of the variance (how you have done if every pot would have been solved as your equity), but I wouldn't recommend that for your analysis in spins, as it doesn't take multiplier variance into account at all. EV Chips will take also multiplier variance into account and that's what I would use as a measuring stick.

 

That is the additional variance in the games. Because prizepools are drawn, you might have really good runs with regards of wins, but still lose money over the period. You see it the other way as well, on the 7$ games for you, you are losing ev adjusted chips, but your Net $EV would be about 8$ in the positive. So in sevens you have ran very well compared to what you should have in the prize pool draws and in the 3s you have had the opposite run.

When I'm comparing the net won chips and ev adjusted, likely there is a fluke in your ITM. Really small samples though.

In general, just counting from 177 games and 5180 EV adjusted chips, with this data probably the expectation would be like losing with negative 1-2% ROI. Not winning enough chips to win after rake takes it's share. Really small samples though, so even though that is the most likely explanation, your game may still be -12% or +10% in the long run. There's still a lot of uncertainty even when looking at the EV Adjusted chips when we're talking about just less than 200 games.