Comparing Simple Strategies

Yikes… this takes a bit longer than I’d hoped the way I did it, so I’ve only finished the flop. I’ll try to post the turn and river numbers a bit later. They’ll be a bit easier, as there is less volume (both due to my own folds, and when my bets generated folds).

Total flop statistics:

  • Folds (60): 23 1F, 24 2F, 11 3F, 2 4F
  • Checks (89): 40 1X, 23 2X, 13 3X, 8 4X, 4 5X, 1 7X
  • Calls (26): 11 1C, 10 2C, 5 3C
  • Small bets (25): 16 1s, 8 2s, 1 3s
  • Half pot bets (37): 25 1h, 6 2h, 5 3h, 1 4h
  • Normal sized bets (7): 6 1n, 1 2n
  • Pot sized bets (8): 3 1p, 2 3p, 1 4p, 1 5p, 1 6p
  • Big bets (1): 1 2o

Note that against a single player, I have 50 bets to 40 checks to 23 folds to 11 calls, while in multi player pots that becomes 28 bets to 49 checks to 37 folds to 15 calls. I’m forced to be somewhat more passive against the enemy “team” than against one opponent.

Oh, and one final reminder: for each street I only recorded my final action, with the number of opponents left when I took that action, and so some of the bets and folds both may have been preceded by a check, for example. I remember one bet, just recorded as “1h”, which was actually me making a small continuation bet into 4 opponents, one opponent raising, everyone else folding, me firing a pot sized re-raise, my opponent raising over that, and then me jamming all in over that (which at that point was only a half pot bet). So some of the time, this can also under represent aggression.

1 Like

Turn statistics for Hard Rain:

  • Folds (30): 15-1F, 9-2F, 3-3F, 2-4F, 1-5F
  • Checks (60): 40-1X, 12-2X, 5-3X, 2-4X, 1-5X
  • Calls (16): 11-1C, 3-2C, 2-4C
  • Small bets (8): 16 1s, 8 2s, 1 3s
  • Half pot bets (21): 17-1h, 1-2h, 1-3h, 1-4h, 1-5h
  • Normal sized bets (9): 4-1n, 5-2n
  • Pot sized bets (12): 8-1p, 2-2p, 1-3p, 1-4p
  • Big bets (5): 4-1o, 1-2o

Again, looking at heads up play: 49 bets to 40 checks, to 15 folds, to 11 calls. Multi-way it becomes dramatically more passive on this street: 14 bets to 20 checks to 15 folds to 5 calls.

Note that all of these statistics are not limited to times I had been the pre-flop aggressor. It might be nice to split that out from the lines where I limped or called pre-flop, where I expect my play in general was at least somewhat more passive. I think in a future version I might look at that also, as you’re likely to want different targets for the two (given that most players donk much less frequently than they fire a continuation bet, for example).

One other thing I notice when comparing turn play to the flop play: I seem to have made a large shift on the turn away from small bets. On the flop, nearly all bets were small or half pot, while on the turn, half pot and then pot sized bets became more common. I think I could have increased my aggression frequency by just replacing some of the turn checks with small bets, but I will often move from a small flop bet to a larger turn bet, both with value and with bluffs. In part, this can be because of a turn card that creates greater equity for a big part of my opponents range, but even more simply: if a small bluff didn’t work, “well, let’s try a bigger one”; and on the opposite front, with a value hand, “well, he had a hand that could continue with a small bet, so…”.

And finally, river stats for the Hard Rain run:

  • Folds (12): 8v1, 2v2, 2v3
  • Checks (36): 26v1, 7v2, 2v3, 1v5
  • Calls (17): 14v1, 2v2, 1v3
  • Small bets (6): 6v1
  • Half pot bets (13): 12v1, 1v2
  • Normal sized bets (1): 1v1
  • Over-bets (1): 1v1

Notice that there is not nearly as much multi-way action on the river, as prior streets mostly thin the field.

I’m a little surprised to see how passively I played the river, with folds and calls (29) significantly outnumbering bets (21). I also think I somehow drifted into smaller sizing than I’d normally like to see, with half pot bets really dominating, and almost nothing larger than that. Sometimes that’s just a natural result of the boards and the range matchups, but if this had been a longer run, I’d be likely to feel I need to find more spots for larger bets.

I also get the sense that I may have been under bluffing on the river, although that doesn’t feel like as large a leak initially as the lack of larger bets. But again, I think it is typically best to find the right combos in your range to make various plays with, and those often come in strange clusters.

Results summary for all runs:

  • Hammer @ 1/2: 222 BB/100
  • Super Nit @ 1/2: 856 BB/100
  • Passive Fish @ 2/4: 73 BB/100
  • Lazy Limper @ 2/4: 372 BB/100
  • Value Village @ 5/10: 220 BB/100
  • Pre Flop Hammer @ 5/10: 241 BB/100
  • Value Village @ 10/20: 176 BB/100
  • Lazy Limper @ 10/20: 244 BB/100
  • Pressure Cooker @ 25/50: 139 BB/100
  • GAG @ 25/50: 300 BB/100
  • Robo TAG @ 50/100: 131 BB/100
  • PFP @ 50/100: 130 BB/100 (guest run)
  • GAG Cook @ 50/100: 190 BB/100
  • Pre Flop Hammer @ 100/200: 2 BB/100
  • Get da Fishy @ 100/200: 104 BB/100
  • LAG Problems @ 100/200: 125 BB/100
  • PFP @ 200/400: 30 BB/100
  • Orphaned Pot Adopter @ 200/400: 49 BB/100
  • Value Village @ 500/1,000: 109 BB/100
  • Balance @ 500/1,000: 153 BB/100
  • Balance V @ 1k/2k: 11 BB/100
  • The Maniac @ 1k/2k: 39 BB/100
  • Light Rain @ 2k/4k: 14 BB/100
  • Random Fool @ 2k/4k: 26 BB/100
  • Multiple Personality Disorder @ 5k/10k: 66 BB/100
  • Balance B @ 5k/10k: 109 BB/100
  • In Pairs We Trust @ 10k/20k: -15 BB/100
  • Small Ball @ 10k/20k: 3 BB/100
  • Super Nit @ 20k/40k: 58 BB/100
  • Laggy TAG @ 50k/100k: 27 BB/100
  • Hard Rain @ 50k/100k: 42 BB/100
  • Night Rain @ 50k/100k: 77 BB/100

New run started on 20k/40k that I’ll be calling Donkey Fever. It involves some twists on opening play and play on the flop, that might possibly handicap me a bit (it will be interesting to see how much), but will feature mostly my normal game on the turn and river.

Donkey Fever run completed:
Tables played: 20k/40k NL Holdem 9 max
Hands played: 1,004
Chips won: $11,231,000
BB/100 hands: 27.97

Here’s a hand very near the end with my getting incredibly lucky:

Here’s almost the last hand played, with luck reversed (although not nearly as bad a beat as above):

The strategic aim here was to try out a high frequency donking range on boards somewhat less favorable to the pre-flop raiser, or on wetter, scarier boards. To give me more opportunities to donk than I’d usually have, I resolved to turn my normal opening range from seats 6 through 3 (everything before highjack) into an open limping range. From there, I divided my range into 4 segments:

  • the strongest segment I would mostly limp 3! (roughly 60 to 75% of the time depending on specific conditions), and then flat the rest
  • with the second strongest block, I would 3! most high off suit holdings, and flat the rest
  • with the 3rd block I would mostly flat, but fold a small amount if the situation seemed really unfavorable
  • with the bottom 4th, I would mostly limp fold

The test run didn’t really work out, in that I think I might not need the fingers of both hands to count the number of times I limped early and then got raised. I’d guess it was 5 or 6 times in 1,004 hands. I did get to call raises from the blinds, which gave me one other set of donking opportunities, but I didn’t get nearly as many samples as I was hoping for, and so the majority of the run was just me playing my normal game, if kind of on auto-pilot. I would expect I lost a bit of EV by open limping from all of the early seats… I had several AA-TT type holdings where I won very small pots, and would have probably won much bigger pots with at least a few of them by open raising.

In the spots I did donk, I was suprised to get almost no folds. I’m imagining that many of the raises reflected very tight opening ranges, and so if a lot of the hands were AA-JJ and AK, you can see people often getting wed to those, and under-folding. I had gone into this thinking donk bluffing might be fairly powerful, but from the limited sample I had, I’d now conclude the opposite, that you might want to donk mostly with some of your strongest holdings, at least into a similar player pool.

But again, the sample was really too small, and I think different players will respond very differently, also.

You might want to use a bit of free software called OBS Studio, which allows you to record anything on your screen. You could then play it back after the session, with pause/play/rewind/fast forward controls that would let you collect the data in your own time instead of as you play.

Yeah, this would take longer, but it would let you collect all the data you want.

1 Like

Results summary:

  • Hammer @ 1/2: 222 BB/100
  • Super Nit @ 1/2: 856 BB/100
  • Passive Fish @ 2/4: 73 BB/100
  • Lazy Limper @ 2/4: 372 BB/100
  • Value Village @ 5/10: 220 BB/100
  • Pre Flop Hammer @ 5/10: 241 BB/100
  • Value Village @ 10/20: 176 BB/100
  • Lazy Limper @ 10/20: 244 BB/100
  • Pressure Cooker @ 25/50: 139 BB/100
  • GAG @ 25/50: 300 BB/100
  • Robo TAG @ 50/100: 131 BB/100
  • PFP @ 50/100: 130 BB/100 (guest run)
  • GAG Cook @ 50/100: 190 BB/100
  • Pre Flop Hammer @ 100/200: 2 BB/100
  • Get da Fishy @ 100/200: 104 BB/100
  • LAG Problems @ 100/200: 125 BB/100
  • PFP @ 200/400: 30 BB/100
  • Orphaned Pot Adopter @ 200/400: 49 BB/100
  • Value Village @ 500/1,000: 109 BB/100
  • Balance @ 500/1,000: 153 BB/100
  • Balance V @ 1k/2k: 11 BB/100
  • The Maniac @ 1k/2k: 39 BB/100
  • Light Rain @ 2k/4k: 14 BB/100
  • Random Fool @ 2k/4k: 26 BB/100
  • Multiple Personality Disorder @ 5k/10k: 66 BB/100
  • Balance B @ 5k/10k: 109 BB/100
  • In Pairs We Trust @ 10k/20k: -15 BB/100
  • Small Ball @ 10k/20k: 3 BB/100
  • Donkey Fever @ 20k/40k: 28 BB/100
  • Super Nit @ 20k/40k: 58 BB/100
  • Laggy TAG @ 50k/100k: 27 BB/100
  • Hard Rain @ 50k/100k: 42 BB/100
  • Night Rain @ 50k/100k: 77 BB/100
1 Like

OK, Sklansky Hammer coming up before too long. I haven’t decided on the stakes yet (though I’ll avoid stakes I’ve already played Pre Flop Hammer, since this is just another variant of that), and I haven’t quite decided on the final formula to determine shove ranges. I do think I’ll do a min buy in, just to maximize the extent to which the shove range is likely to change. I’ll probably also use effective stacks in the formula, rather than just focusing on my own stack depth.

Credit for this idea goes to SunPowerGuru.

2 Likes

I finally caught up lol

I wonder if it isn’t a good idea to start with the proposed theory exactly as proposed and then use your experience to refine it?

I don’t know if we can call it Bayesian but that’s kind of what I’m getting at.

Regards,
TA

edited to add:

The discussion began in another thread here:

As described I felt I just wouldn’t be going all in enough. I’m most of the way through the run finally, and even going all in much wider, this has been more boring than I remembered. It’s funny how mood works, as I don’t always find this so painful, but this run has been taking a while just because I often haven’t been able to bear with this for more than 100 hands or so (and often even less), and so it is running into multiple days to finish.

I’ve been playing 25/50. The run looks likely to finish up a solid amount, and so you’d think that might make it more bearable, LOL. So far it seems to be an example of how you can certainly win money on the low stakes tables with a shove or fold only strategy. If you want to try it, I’d suggest watching a movie or reading a book while you’re doing it…

2 Likes

You say, “…win money at the low stakes tables.” Where does that come from? Have you tried it at higher stakes?

I see this a lot on this forum, the “that would never work against the super elite players at the tables I play” attitude. I have news for you… a lot of it does work at all levels.

As an example, Sklansky’s system was tried by a casino owner, and it took him to a final table in a $2k buyin WSOP event.

Statements like that need to be backed by data, not just ego. How would Replay’s super elite counter this strategy when some of the world’s best real money players could not?

Well, the statement that it wins at low stakes is backed by data. I’ve done 3,000 hands now, roughly, and while the current run is not finished, it appears likely to finish with a winning run also.

I had one 1,000 hand run with a Pre Flop Hammer variant at medium stakes (with a fairly Skanksky-ish approach), at 100/200, and it finished up also, winning 2 big blinds per 100 hands played (and I think I probably ran bad during that run). I have seen people playing mostly shove or fold strategies at elite stakes before, and it has seemed like it has not performed horribly. My prior assertion that it performs well at low stakes was just that: I’ve got enough data to assert that it performs well at low stakes.

Do I think it would perform well against strong players? No, I don’t. Do I have data to back that up. Nope, I’ve never wanted to try as I didn’t want to throw my chips away. But at the same time I think it would probably perform better than a lot of the strategies I see people trying. It seems to me likely to be sub-optimal on a number of fronts, but in an aggressive environment, with a lot of pre-flop raises, it can actually start getting more effective again.

As for the casino owner that got to a final table with this strategy… I’ve noticed that final tables are not monopolized by professionals. There’s a lot of luck getting to a final table. I notice you’re not asserting that this casino owner got to a final table with this strategy multiple times. Nor that hundreds have followed in his footsteps, becoming millionaires through the power of the Sklansky formula. So I’m not sure that constitutes strong evidence that this is an effective strategy against strong opponents, either.

So how good of a strategy is it? I’d say it is surprisingly good. It has crushed every run I’ve tried with something similar at low stakes, and I’ve tried it at medium stakes expecting it to be a winner. Like any strategy (aside from truly perfect play), I expect it will be pushed -EV at some point, and I’m not exactly sure where that point is. Since it is one of the simpler strategies I’ve employed in this post on simple strategies, it’s easy to assume it’s probably one of the weakest, and that might not be the case.

If anyone is successful with it on a high stakes or elite stakes run of 1,000 hands or more, let me know.

I should also note that I don’t mean to be dismissive of the strategy just because I’m not enjoying the run.

First, I’m not playing Sklansky’s strategy, I’m just doing a run inspired by it. Second… mood is a funny thing. I’ve definitely had times in the past where I’ve enjoyed playing in a style like this… lately I just haven’t been in the mood for folding a high percentage of hands and just shoving with the rest. But what is boring for some is relaxing and peaceful for others.

A follow up on Sklansky Hammer

There are a number of things that can make strategies like this hugely effective:

  • people making very large pre-flop raises with wide ranges
  • getting wide calls when you move all in

I’ve been seeing both, but especially the latter, in this latest test run. Some hands I recall getting calls with: A5, Q5s, 87s, KT, QJs, 84, AA, TT, JJ, AQ, AT, AQs, 88, 77, AJ. Some of those hands clearly just don’t belong in the mix calling a 50BB all in raise (and in some cases a 200BB or larger raise). The more of these random calls you get, the more impressively a strategy like this will work.

Yorunoame, I humbly apologize. When i read that this strategy would work at low stakes, my head added “but it would never work at the stakes I play,” and I kinda snapped. You never said that, you have never expressed anything like that, it was sloppy and wrong of me to assume that.

For more than 5 years, I have seen that kind of thing said over and over, ranging from “90% of the players here are idiots,” to (literally) “that would never work against good players.” I’ve seen the chilling effect this has had on the community. I don’t know how many people have stopped posting here because of that type of comment, but it’s a lot.

I have no attachment to Sklansky’s strategy. I don’t like it because it was designed to replace skill and experience. It cheapens the game, in my opinion. I remembered seeing it years ago, and mentioned it in response to a comment about “bingo” play, which is another common theme here.

I have been following this thread with great interest. I appreciate the time and effort it has taken. I think it’s a valuable addition to the forum.

My snarky comment was uncalled for. It was based on incorrect assumptions, and was just wrong in many ways. There’s nothing left for me to say except I’m sorry.

5 Likes

LOL… I re-read my initial post with a little horror, also. I had just hoped to make a post where others could laugh at my suffering. Sometimes I love doing these runs, but other times I’m just hoping somehow I’ll magically get to the 1,000th hand without my really having to sit there and play it. And I’m not always sure why it is psychological torture sometimes, and pleasant and relaxing other times.

3 Likes

Said like a true gentleman, bravo SPG. Much respect.

2 Likes

OK, just had a bit of a bad stretch, and lost much of what I’d been up. Down to about 100 hands more to play.

The hands that called all in bets during this session and showed (I was often covered by others, and so didn’t see hands in multi-player pots that were not put all-in and did not win): A4s, K8s, AJ, T8, TT, K7, 74s, JJ, AQs, AA, 73s, QJs, KK

Given how weak calling ranges have been overall (I think the cards above are probably the strongest I’ve seen over a comparably sized sample so far), I think in a future run (if I did one) I’d modify my yet to be revealed ranges so that these very tight ranges are slightly wider when I have a high key value (large effective stacks and/or many possible opponents), and quite a bit narrower when I have a low key value (smaller effective stacks and/or fewer possible opponents), so that I’m just using a more linear range that prints chips with every shove. The basic idea of going tighter when stacks are deeper and when you have more possible callers, and widening rages with shorter effective stacks or with few opponents is valid, but I over did on both ends in this run, with this particular pool of opponents.

I’ve also been doing the min buy in every time for this run, which has helped create a wide range of key values, and thus a lot of variety in the ranges I’m shoving with, but a drawback of this is that it creates more volatility in the run. Starting with max buy-in means that each shove is risking an amount that is closer to the same, while starting with min buy in often means you can win multiple shoves in a row, but then still lose everything in a subsequent shove. Once you are the big stack at the table, each shove is closer to a linear adjustment to your bank roll.

Over a long enough sample this wouldn’t matter, and having the diversity can make it harder for opponents to accurately guess the shape of your ranges, but over a 1,000 hand run it increases the bumpiness of the ride. It occurs to me that the only other run with a pre-flop shove formula that did not do well also involved a minimum buy in.

Anyway, this may still do well, but the last 150 hands or so were a bit of a disaster.

1 Like

Ick… disastrous finish. The last 100 hands were worse than the prior 150. I’m pretty confident this would be a winning strategy at this level, even as I played it, and later in this post (along with the prior post) I cover some things I think might help improve EV for the idea.

Sklansky Hammer results:
Tables played: 25/50 NL Holdem 9 max
Hands played: 1,002
Chips lost: -$2,129
BB/100 hands: -4.25

I was probably running better than 60 big blinds per 100 hands over the first 750 hands… so the final 250 hands constituted a pretty bad patch. I think I might have lost over a dozen shoves in a row at one point, getting chips in ahead on all but two of those. It’s not actually the case that I had even 50% equity on many of those, though, as quite a few were multi-way. On the two I was behind, my equity was quite poor, so maybe my average equity over that stretch was roughly 40% or even less (4 way, even with the best hand, 30% equity is pretty good). But even so, losing 12 in a row is not what I’d call running good. But so it goes.

Mechanics used:
For simplicity, base was just effective stack depth in big blinds. Since I was doing a min buy in of 50 big blinds and was usually the small stack, that meant for the first few hands around the table, my beginning key value was usually in the 40 to 50 range.

To modify that by the number of players I might face, I chose to generally weight all players that could still potentially call my all in bet equally, whether they limped or were left to act behind, and even often if they raised. Many players were limping with close to 100% of their hands, and so a limp may have made their hand weaker, just by reducing the likelihood of a premium holding. Even raises I often ignored, especially min raises from players that min raised pretty much every hand. If a given player did not limp often, or if there was a raise from a player that did not often raise, I weighted that player double. A 3! was usually counted as 4 (unless it was from a maniac that was always raising), and so 2 limpers, a raise, a 3!, and 1 player to act behind would be (2 + 2 + 4 + 1) for a total of 9. If I had 50 big blinds, that translated into a final key value of 450 (and so would have shoved with all of the hands through the 250+ row).

Hands shoved by key value:
4,000 +: AA
2,000 +: add KK
1,000 +: add QQ, AKs
500 +: add JJ, AK, AQs
250 +: add TT, AQ, AJs - ATs
100 +: add 99-88, SC → T9s, A5s - A3s
50 +: add 77, AJ, Axs, SC → 54s
25 +: 66-22, Ax, S1G

Again, I think the ranges in the last 3 are too wide (especially the last 2). I also speculated that the ranges might have been too tight on the upper most end, but I think that is a bit closer call. I got a stack of over 400 big blinds on a few occasions, and if I faced 6 or 7, even with no meaningful raises, that would push me over 2,000, where I would only shove with AA and KK, and I suspect QQ, JJ, AK and AQs are still profitable shoves as wide as people were calling.

I should also mention that I did flat a few times out of the small blind, and in those situations and as big blind, I allowed myself 3 options post flop: fold, call or shove.

Edit: I also think that people that try this but stick to the original Sklansky formula and shove ranges would see less variance than I did, on average. I wasn’t too worried about variance, but if you tried this with a bank roll of less than a million, you might consider that route.