The use of databases for hand selection is very interesting. I suspect it is most useful for RFI ranges and 3-bet ranges against RFI and probably provides a good starting point. However, how useful is it in spots which are presumably are less frequent in the database such as a lot of limp/calling and flat calling action? Also, the database results are by its nature averaged over all player styles and abilities unless you can filter for those as well (which would perhaps reduce the sample size considerably depending on the number of constraints). I’m curious to hear more about that.
Are any of these databases and analysis tools freely available?
That is a key question in the context of this site (and I assume recreation poker in general), where limping in and limp-calling is the predominant action, at least up to the 20k/40k level.
I was planning on asking Yorun if his pre-flop ranges depended on if the action folded to his seat or if it accounted for pre-flop limpers. For example, what does the opening range look like in MP with three limpers in front of you and four players yet to act?
Of course, this is an incredibly complex question as it needs to factor in what players tend to limp in with. For most players, their RFI range is JJ+ and sometimes AK at most, and I have seen many players limp in with KK, QQ, JJ, and AK on a regular basis.
This complicates things in terms of constructing your range and bet-sizing in MP, since a limper could either have a premium hand or a marginal hand. Their passive play inadvertently disguises their actual hand strength.
This connects back to Yorun’s follow-up post on c-bet ranges. Based on the pre-flop action, there are many spots where it will be difficult to know your relative hand strength, particularly if 3+ players have limped-called your open. I actually find MP the most difficult to play because of this.
Personally my open range is dependent on limpers.
I have often made the mistake of opening my bottom range after limpers, and I recently learned from a French poker player that after a limp I am supposed to open with better hands than my usual range (gap concept).
Basically in limped pots I have a specifically tighter open range and it has worked well for me over the last 2 months since I learned this.
For example on a 6 max table a hand from my bottom range UTG+1 is ATo which I open when UTG fold and which I fold after a limp.
Off topic by the way but I also have a limp range which also depends on the pot if it is limped or not, and unlike my open range the more limpers there are the wider my limp range is
As an example in cash game 6max in a non limped pot I limp between 2 and 3% of my hands (mainly pocket pairs) and with one or more limpers (and also depending on my notes on the players I admit) I limp up to 12% of my range, and 16%-20% yesterday against a very particular player, in short nothing GTO
Limpers in front in general should probably result in a tightening of your opening range, but there are spots where the opposite can happen. The most common example is where one of the limpers (or a player still left to act behind) is a particularly tasty variety of fish, then raising in hopes of isolating on the fish can actually result in widening ranges.
What do I do typically?
who is limping?
some players rarely open limp except with premium holdings like AKs, KK and AA; if one of these players is in the mix, I’ll tighten my opening range significantly
some players never raise pre flop; here, as premium holdings are still in their range, I’ll tighten my range by moving my effective seat forward by one seat
with limpers that sometimes raise also, the limp has fundamentally left them with a weaker range, and this tends to be the most common scenario
who is left to act behind?
if I have passive players behind me, I will start limping behind with a broader range of speculative hands
if I have more active players behind me, I will tighten both my limp-behind range and my raising range
With limps in front of me, on the whole I prefer attacking: any folds generated are free chips. So by default, I typically broaden my range slightly, but break it into two parts:
I take some of my normal range that is more speculative (small pairs, Ax suited, suited connected cards), and limp behind with about 60% of these hands, and also limp behind with all expanded parts of my range that would not normally be an open from that seat
I raise with the rest
Edit: oh, that addressed limpers, but not other actions
with a raise in front, my combined call or 3! range is typically much tighter than my open range for the same seat
with a raise and cold call in front, my 3! range will expand, and my calling range will typically contract slightly
with a raise and 3! in front, I’m normally going to be silly tight… no one 3 bets on this site, and so unless it is a player with a higher 3! frequency, 4! and cold call ranges will usually consist of either AA or AA + KK. With stronger players, that will widen dramatically.
Almost all ranges are for RFI / vs RFI situations rather than for open limping or vs limps. There aren’t many games where players open limp a lot. I’ve seen some material on this from a few coaching sites and they are simply adjusting off a standard strategy depending on the type of limper(s) in the game. I think this is the best way of adjusting your standard ranges because not all limpers are the same. You have nits and loose passives and some trickier types mixed in. IMO, player specific adjustments need to be made on the fly, not incorporated into your baseline range construction. I could be wrong here if the predominant action in your games are limps though.
Databases can be filtered by stakes and even by sites to give more accurate population reads. The sample sizes are all still large enough to be very accurate. I know of a training site that compiled its hand profitability matrixes over a sample of 600 million+ hands. Since all these hands were played by students and the site is tailored to entry-level through mid stakes cash games, those are numbers I’d have high confidence in.
I don’t think there are publicly available databases. Data is extremely valuable in poker and people either keep it private or try to monetize it. People are happy to sell access to the ranges they have created from the data they own but won’t give you access to the raw data itself. I will look around to see if there is anything in the public domain.
I would suggest taking advantage of whatever free resources you can find from reputable training sites. pokercoaching.com has a free membership to give people an idea of what is behind the paywall. There is a lot of content on ranges and range construction you can look through for free. pokerheadrush.com also has free content specifically for MTT players. I will look for free tools. pokerstrategy.com has equilab, which is wonderful for evaluating hand v range(s), range v range(s) - pre and post flop.
I’m sure there are other effective approaches, but I just wanted to second this. I mostly divide open limpers into 3 categories that then feeds into on the fly modifications of my raising range:
players that predominately raise strong hands and limp weak hands: these players create an incentive to widen my raising range (in much the same way that a cold call creates an incentive to squeeze), as they will often limp fold, and it is also nice when they limp call
players that play a nearly 100% limp strategy: these players often have premium hands in their limping range, but I still won’t tighten my range to the point of my 3! range, as their range is still generally wider than an open raise from the same position; I tighten my raising range slightly
players that don’t frequently limp, and will often limp 3! (particularly from early position) as a trap with premium holdings: I tighten my raising range almost as much as I would as if there had been a raise from that seat
Here, I think most limping falls into the first category, and encourages me to raise with a range wider than I’d raise with if everyone in front of me had folded, although I don’t want to exaggerate the effect of that, as some of my open raising hands also find themselves converted to limp behind hands. If my RFI in a given seat was 18%, I think my range with weak limpers in front really looks something like 16% raise, 8% limp behind. So my VPIP goes up, but my actual raise frequency goes down. But at the same time, the weakest hands I’m raising with have grown weaker.
It’s not always quite that simple… but those are at least significant considerations that modify my pre flop range construction with limpers in front.
OK, I’ve done some research and am getting a reasonable approach together. In the meanwhile, this is something by Evan Jarvis on the foundation of range construction and some of the key concepts:
