Know Thine Enemy – Profiling Your Opponents For Poker Profit

Hollywood has given many the impression that expert poker players make their fortunes by correctly interpreting a pulsing neck vein, or by peering like vampires into the souls of their victims and running mind-bending bluffs. In contrast, nearly all successful players I know agree that the biggest jump in their expertise stemmed from developing their hand-reading ability. The importance of hand reading is readily apparent here at RCP through multiple videos, quizzes, and other products, and rather strangely confirmed in an e-mail from RCP contributor Mike Gano that I received while I was editing this article.

As Vanna White will tell you, you cannot spell “hand reading” without R.A.N.G.E. Systematic hand reading is based on assigning opponents ranges and then pruning them as the hand develops across multiple streets. It was the ranges given in Ed Miller’s book “How to Read Hands at No-Limit Hold ‘Em” that eventually led to my realization that my laziness was, once again, costing me money.

To put this in context, wind back the clock to when I was an online LHE grinder marooned in Kansas. As I mentioned in my previous article, PokerTracker allowed me to eliminate a lot of leaks from my game. However, like most people I primarily used such software, along with a HUD, to profile and thus exploit my opponents.

Whenever I flew out to Vegas to play live poker, the first thing I noticed was the absence of red and green numbers hovering above my opponents’ heads. It would usually take me at least a session to get into a live zone in which I could develop general profiles for those facing me across the felt. I’d mark the ancient gentleman in the 1-seat as a nit and the young drunk in the five a bad LAG and move on from there.

When I finally got around to taking hand reading seriously, I realized these general profiles needed to be translated first into starting hand ranges. I was delighted to discover that Ed had done this work for me in his book. But then… Again, I’ll walk you through this by retracing my footsteps.

Ed’s range for The Nit in “How to Read Hands” is:

AA-22
AKs-A2s, KQs-KTs, QJs-54s, QTs
AKo-ATo, KQo-KTo, QJo-JTo

When I first saw this range I blinked at it. I stuck it into Equilab to see if if looked more plausible when displayed graphically. I shook my head.

“Ed, baby,” I said to myself, “I dunno where you’re playing, but that ain’t a nit.”

I ran the mouse over the Equilab grid, popping out offending cells and settled on:

AA-99
AKs-ATs, KQs
AKo-AQo

“Now THAT is a nit,” I announced confidently.

In fact it’s two nits that I play against regularly. Seriously. They play that tight and fork the range so that the weaker hands are limps.

It was at this point that the story got interesting for me and hopefully useful for you. Because it occurred to me that I had been placing a nit lanyard around the necks of many of my opponents, particularly older Vegas regulars, and adopting a similar strategy against all of them. And while I was confident that my “nittiest” range given above was applicable to a couple of those individuals, it also occurred to me that other members of the nit pool likely played wider. But by how much?

The “Know Thine Enemy” project was born.

I’ll get to the “how” of this project in a moment, but first I think it’s helpful to summarize some of my findings to justify why you might engage in similar work.

Given my online experience, a natural way for me to characterize my opponents is through the two critical numbers of VPIP and PFR. More generally these numbers provide a simple metric for how loose/tight (VPIP) and aggressive (PFR) players are pre-flop, and lead to decent approximations for the ranges we need for hand reading.

This is how it works. My range for the two super-nits implies they play 6% of their hands. This is effectively their VPIP. If I determine that a specific opponent has a VPIP of 12% I can play with the whammy bar in a program like Equilab to approximate their opening range, and then leap into all the entertaining and useful hand-reading exercises provided by RCP coaches.

So I started collecting VPIPs of my regular opponents, many of whom I had mentally labeled as nits, expecting to get values somewhere in the 12-15% range, and certainly tighter than Ed’s “The Nit” range which corresponds to a VPIP of 21%.

I had hours of fun with this. There’s Ol’ Walker coming in nicely at 14% pretty much identical to Little Denny, and here’s Silent Dave at 18%, well now, bit of a surprise, I thought he was tighter than that, and WHAT THE HELL IS THIS?

Sleepy George 27%. Crimea 30%. The Cardinal 36%!

I was devastated. I thought I was decent at paying attention at the table and that I had a pretty good grasp of how tight my regular opponents played. And yet someone I had marked as a nit was splashing about with 36% of their hands.

If you conclude at this point that I’m simply an idiot, I wouldn’t blame you in the least, but please hang with me for a bit longer. When I shared these data with people I respect who are also familiar with these specific opponents, they were also shocked by the high VPIPs I was getting. One was kind enough to do some data gathering and confirmed my results.

You might reasonably object that anyone playing more than 30% of their hands would reveal the fact by showing down junk. I have no doubt that on occasion they do and in some cases I’ve simply overlooked these hands or attributed their presence to an anomaly possibly driven by tilt. But what additional research revealed is that the reason I wasn’t seeing these weak hands and being alerted to how wide these players were entering pots was because they were not showing them down. The only explanation was that the vast majority of the time the junk hands were being folded postflop.

Many of you may now be salivating at the thought of such opponents and the obvious and simple ways to exploit them. This is precisely why it is so valuable to identify them.

To further defend my apparent lack of observational skills, I suspect a bunch of cognitive biases are at play, combined with the possibility that without specifically recording frequencies of events requiring a baseline of at least tens of hours, our brains are not very good estimators. If true, it’s possible that you’re making similar mistakes.

There’s also a technical point which may be the only original contribution I will ever make to poker theory and which I have dubbed “limit residue.” To illustrate it, let’s return to Ed’s “The Nit” range. If you display this graphically on your favorite scenario analyzer you’ll notice immediately that the range is asymmetric in the sense that the suited hands are favored over the unsuited ones. This is, of course, a feature of any reasonable NLHE range.

The superiority of suited to unsuited hands is less marked in LHE. In fact back in the online heyday of LHE, if someone won a pot with 74s you could guarantee the chatbox would light up with “You play that crap?” immediately followed by “But they were sooooooted,” “LMAO” and “STFU.”

Many of the older $1/2 NLHE regulars in Las Vegas and elsewhere spent decades playing LHE and I am convinced many of them are covered in this “limit residue.” They greatly underestimate the importance of suitedness in NLHE. And as a result, their ranges in the 13×13 grid will be much more symmetrical than is ideal.

If you take “The Nit” range and symmetrize it by giving each suited hand its offsuit partner, the associated VPIP climbs from 21% to 35%. The irony here is that you rarely get to see those bad offsuit hands shown down, particularly from overfolding Vegas regulars, precisely because they play so badly in NLHE.

So how do you measure VPIP at the table?

Just find something that works for you. There are phone apps that will track VPIP/PFR that just require the user to tap a box whenever a player calls or raises pre-flop, although personally I find them awkward to use. Players leaving or walking complicates matters to the point it takes my head out of the game. My preference is to identify one regular per session and to track their actions in my phone’s notes feature. This is much less distracting than tracking multiple players and since I see regulars day after day I have ample time to build up a solid profile. If you’re concerned an electronic device might contravene local gaming regulations, just use pencil and paper.

Another advantage of tracking a single opponent is that I have plenty of time to also record their further actions in a hand. Obviously whenever they show down a hand I make a note of it and the position from which it was played.

Against tourists who I am unlikely to face repeatedly I revert to relying on their worst shown down hands to estimate their likely ranges. Clearly all you can ever accomplish with this method is to set a lower bound on their VPIP, but this is still useful, particularly since it alerts you to loose, exploitable players fairly quickly.

The idea of limit residue has a bearing on another important point: there is not a direct mapping between VPIP and range. In addition to the degree of asymmetry between suited and unsuited hands, players assign different relative values to pairs, broadway couples, suited connectors, and so on. For example, if you plug in 25% to Equilab it spits back all pairs down to 66 and offsuit aces to A7o. In reality most $1/2 players who will enter a pot with 66 will do so with all pairs. Similarly if a typical low-limit player thinks A7o is a reasonable starting hand they will include all aces, so for a given VPIP tweak the ranges accordingly.

If you’re skeptical that such work has a sufficient pay-off, consider the following. The fact you are at RCP strongly suggests you value RCP content and, by extension, you will have spent some time carrying out hand-reading exercises; possibly a great deal of time. I have no doubt of the value of such work when “assuming a typical range for an average $1/2 regular.” But if instead of making general assumptions you do the same exercises with a well-determined range specific to Sleepy George, then being sleepy at the table has become the least of George’s poker problems.

This article is dedicated to the memory of Louis, ?-2017.