To give some more numbers to help with achievable hourlies (including rakeback per hour). As a rough ballpark for hands / hour at 6-max normal speed (not fast tables)

From a very unscientific observation
NLHE will deal approx 75 hands / hour
LHE will deal approx 100 hands / hour
PLO will deal approx 55 hands / hour

The VPP / hand numbers used later in the post were sourced from fpppro.com and are based on 6-max.

So an average winning player produces the following $ / table / hour + rakeback. Rates in parentheses are (bronze|silver|gold|platinum|supernova).

NLHE (ptbb / 100)
=================
NL2: $0.14 + 2.5 VPP ($0.03 | $0.04 | $0.06 | $0.08 | $0.14)
NL5: $0.20 + 4.6 VPP ($0.05 | $0.08 | $0.11 | $0.15 | $0.26)
NL10: $0.37 + 7.6 VPP ($0.08 | $0.13 | $0.18 | $0.25 | $0.42)
NL25: $0.75 + 9.5 VPP ($0.10 | $0.16 | $0.23 | $0.31 | $0.53)
NL50: $1.76 + 17.0 VPP ($0.19 | $0.29 | $0.41 | $0.56 | $0.95)
NL100: $2.90 + 27.3 VPP ($0.30 | $0.46 | $0.66 | $0.90 | $1.53)

LHE (BB / 100)
=================
0.02/0.04: $0.07 + ? VPP
0.05/0.10: $0.23 + ? VPP
0.10/0.20: $0.29 + ? VPP
0.25/0.50: $0.74 + ? VPP
0.50/1: $0.86 + 19.8 VPP ($0.22 | $0.34 | $0.48 | $0.65 | $1.11)
1/2: $1.48 + 33.2 VPP ($0.37 | $0.56 | $0.80 | $1.10 | $1.86)
2/4: $3.43 + 51.3 VPP ($0.56 | $0.87 | $1.23 | $1.69 | $2.87)
3/6: $4.39 + 81.6 VPP ($0.90 | $1.39 | $1.96 | $2.69 | $4.57)

PLO (ptbb / 100)
=================
PLO10: $0.32 + 9.2 VPP ($0.10 | $0.16 | $0.22 | $0.30 | $0.51)
PLO25: $0.76 + 13.8 VPP ($0.15 | 0.23 | $0.33 | $0.45 | $0.77)
PLO50: $2.63 + 24.0 VPP ($0.26 | $0.41 | $0.58 | $0.79 | $1.34)
PLO100: $4.61 + 37.6 VPP ($0.41 | $0.64 | $0.90 | $1.24 | $2.10)

For calculations you can use the following conversion rates. I have used the cash bonus rate at that level to work out $/VPP:
Bronze: 1FPP/VPP, $0.010/FPP, 1.1c/VPP
Silver: 1.5 FPP/VPP, $0.011/FPP, 1.7c/VPP
Gold: 2.0 FPP/VPP, $0.012/FPP, 2.4c/VPP
Platinum: 2.5 FPP/VPP, $0.013/FPP, 3.3c/VPP
SuperNova: 3.5 FPP/VPP, $0.016/FPP, 5.6c/VPP

What do I draw from this?
1) At the micros there is little difference between PLO and NLHE on a per table basis. At small stakes PLO appears to start gaining considerable ground in terms of $/table/hour. Note this does not take into account how many tables can be played simultaneously, it is only a per table number.
2) LHE on a per table basis performs surprisingly well even with the poor winrate, especially when rakeback is taken into consideration. Due to the lower edge over the field one must be prepared for extended downswongs compared to NLHE. They are not as severe in intensity as PLO however it is quite possible they last for longer. These downswongs are compensated in part by the very solid rakeback received so as long as you are rolled and are actually a winning player (this part is the key) you can survive.
3) Not in these tables but game selection appears to become a real issue once you pass NL1000, 10/20 LHE and PLO400. After this point you will find lots of regulars and not many tables. This is hardly a problem for anybody in the beginners forum though.
4) Rakeback adds a much larger amount to the bottom line of a winning player than I first thought. Much much larger. If you can add just a single extra table or play for a couple of hours more a month to hit the next VIP level on stars then it is definitely worth the effort.

I am really interested to compare results if anybody is willing to hunt down how winning SnG and MTT players perform.


I never claimed it to be a scientific study. However I did outline how the results were produced so they are hardly "random guestimates" that I pulled out of my arse because I liked the sound of them. I take it you don't like the numbers, perhaps you can add to the discussion with outlining how the results can be improved? Or better yet provide a better sample.

If PLO is less profitable than NLHE then why do people switch at all? It would make no sense for a profitable NL50+ player to switch to a much higher variance game with a lower hourly. Why are you taking issue with the numbers anyway? I went into this without expecting any outcome and without favouring any particular game, I was only interested in what the reality was instead of all the "friends of friends have told me that it is easy to make x playing y" or "I like playing z so therefore I will tell everyone it is the best game". If you can provide a better set of results demonstrating a different outcome I would be more than happy to see them as this was produced mainly for my own information to decide what game to get back into after a break from poker. I posted it here as I often see questions of "what is the most profitable game?" and the answers always give opinions without any quantitative support.

Adding BR requirements of 30BI for NLHE, 300BB for LHE and 50BI for PLO reveals some more interesting information.
I take back my comment to Abe Lincoln, he is actually spot on the money in re PLO not being a good game to move up in. If you have an existing roll and can move into your desired level with a bit of re-training then the returns can be better. However when moving up the higher BR requirement generally means you will be playing a level below what you would if playing NLHE. So if you are playing PLO25 you could be playing NL50 on the same roll. So your $ / table / hour for identical rolls will actually be significantly lower for PLO than for NLHE. Add in the extra fact that generally (there are exceptions to this) the accepted wisdom is that one can play more NLHE tables than PLO and your hourly while moving up for identical rolls can be significantly higher for NLHE. I have not done any comparisons with LHE for identical rolls yet.

More great research in your last post, thanks again.

The bolded parts I found interesting, at least for my online play. I can't play on Stars but have a good rakeback deal on Merge, and looked at my data to see if LHE was a possibility for me. Turns out I can make similar $ as playing 10NL at least on a per table basis. Not sure about the variance yet, not enough data to analyze yet. Merge has very soft Limit tables compared to the NL at micro stakes. Problem on Merge is finding enough LHE tables active, forget table selecting.

I remember a couple years ago I played tons of $.25 to $1 LHE on FTP w/ 27% RB and built my roll so I could explore other games (where I got hooked on SnG's and MTT's) but lately I have been looking at LHE again. Just wish the US had some larger sites to play on, but I'm seeing some possibility of playing LHE online which should help my live play and actually help my BR growth. Somehow seeing the numbers compared side by side makes it OK to play LHE online, at least for me.

I wasnt taking issue with your numbers at all. Just wondering out loud.

This was the answer I was looking for.

denks sir thank you for taking your time to make a nice post like this, though I'm talking to abe when saying I can't figure out his random guestimates that people don't move up in PLO, so relax a bit...

Excellent post OP.

Thanks again, denks. Lots of fascinating stuff here, both in your original figures and in the discussion you've generated.

I guess a study like this can only ever be a snapshot. Going to be interesting to see how PLO fares over the next couple of years. Profitable yes, but bigger bankroll requirement AND - a factor that I'm sure a lot of switchers to PLO or potential switchers to PLO haven't really factored into their deliberations - way bigger potential for tilt.

Also, tons of training vids now aimed at PLO so it'll be interesting to see if those games remain extra-profitable.

Kind of similar maybe, to NLHE a few years back. When I started dabbling in poker, Limit was pretty much everybody's game. Then guys started saying, wow, think I'll switch to NL, it's way more profitable. At which point the vid-coaches and the authors said, well, if that's what you want, that's what we'll produce.

Thanks again.

figures quoted in opening post are ridiculously low

ptBB but they do seem kind of low.

You know the real figures but aren't going to tell us?

people dont tend to play 100k hands at stakes that they are crushing so the numbers are skewed

2ptbb at 10nl is good but not exactly impressive or anything

Yes someone destroying a level will likely move up. However I was trying to get a snapshot of a solid average winning player - something that everyone here in beginners can judge our success by. I was not interested in nor trying to find out "what would Phil Ivey make at NL10?" which would be fun but provide zero useful information to anybody. The question I was trying to answer for myself was not "what is the highest achievable winrate?"

I know the numbers look low but I was not trying to present "impressive" results, I was trying to find what is actually achieved. There is a problem with using people with a smaller sample of hands, they may well have been on a heater or be a table selection nit. Taking someone with a sample of for example 15k hands is next to useless - 7ptbb/100 over 15k hands does not demonstrate anything in regards to what that persons long term winrate is.

I find this to be the most interesting and unexpected result in this snapshot: If a conclusion could be drawn from these figures, it would be that it's much more profitable to play NL50 than NL25, if your bankroll allows. There must be an awful lot of people playing NL50 that can't beat the game if the win rate is actually higher at double the stakes.

I am guessing rake plays a big part in that - rake is comparatively very high in the micros and has a far bigger impact on winrates than people think. I know from personal experience that I had a much higher winrate at 10/20 LHE than 5/10 LHE due very much to the comparative lowering of rake as one moves up.

Yeah, that's a good point. I see what you mean. Sorry - didn't mean to snap.

YEs its a good post, but is it very skewed. Because very few people are going to play 100 k hands at those stakes.

The majority of winning players, have moved up stakes way long before they reached 100 k hands at Nl 2-10.

Your results are certainly biased since you can't filter HU/6max/fullring on PTR.

i kinda understand what the opening post is about now i think, but im not sure what burning question it exactly answers

its ok i do it all the time

All of the discussion regarding players that are crushing a particular level that move up before logging 100k hands can probably be tracked in that anomaly you point out. I don't think NL50 is easier, it's just that it would be far more likely for the data the OP researched to show players who have "settled in" to a level like 50NL and actually stay long enough to log 100k hands. Many of those were probably crushing 10NL and 25NL but did not stay long enough to show up in the OP's data.

I think regardless of the numbers, there is tons of value in his research in that it illustrates the relative winrates given the parameters used, which affect all the lower stakes equally it would be assumed. I agree with DamienT that there is some bias built in to the data because there is no filter for table size which is unfortunate. So I'm not looking as much at the pure winrate data as I am the comparison to other stakes and games in his data.

I am fairly sure there are lots of guys here that could beat the winrates shown in the nano stakes but that is kind of irrelevant to what the OP is showing here. His data does give new members / beginners some reference and a "baseline" of sorts along with a great relative winrate comparison across stakes and games. Which is why this should be a sticky IMO


Cliffs:
1) The winrate data is probably more reliable at the higher levels shown in the chart, since good players will eventually find a more stable level and log the 100k hands specified in the research parameters.

2) The value in the OP's chart is in showing the relative comparison across stakes and games given the common parameters used in the research, so don't focus on the absolute winrate numbers but rather how they compare.

50nl have heads up while 25nl does not.. so yea for sure those #'s are skewed cause people can beat 50-100nl heads up for like 30bb/100 over large samples cause well your bumhunting being h.u with a fish 24/7 i mean its pretty ez game when you do that

50-100nl fr regs winrates usually gets reduced by half or more from 25nl

Bump so it will be preserved in BQ

Interesting data - thanks a lot for the work you did!
I would also guess numbers are pretty skewed for several reasons, still it is pretty interesting. Especially useful to compare conditions between different games, since the numbers are all collected under the same conditions.
I was considering getting into PLO, it's kinda funny to see that winrates do not really differ in the micros...
One question though is this filtered for game size (6max, FR, HU)? Or are they all condensed as an earlier poster suggested?

wow such low win rates seems kind of crazy there wouldn't be more regs on it who settled into those stakes making decent money. Rake is a real killer at those stakes though too.

nice analysis man.

interesting results

although it does have truly good players removed as they will have moved up limits within 100k hands