Reading Poker Tells - Poker Tells Training

Examining Poker Behavior

Live poker tells/behavior analysis

LIVE POKER BEHAVIOR/TELLS ANALYSIS SCHEDULE

Here’s a calendar (subject to change) for some upcoming live poker tells analysis streams. You can watch these on my youtube: www.youtube.com/readingpokertells.com. Sessions will likely last 2-3 hours. Almost all footage analyzed will be footage I’ve never seen before, and I’ll usually be watching with hole cards blocked to eliminate bias (more info on my approach is below the schedule).

  • Dec 13 2025, Saturday, 12 noon EST (recorded). This started with me analyzing a random stream, but after a rough start due to a subpar stream, I switched to analyzing the footage of the Hustler Casino Live game the day before the infamous Robbi Jade Lew J4 hand. You can see the recorded Twitch stream here for a week after recording. This is not a great watch; I was largely getting the kinks worked out format-wise.
  • Jan 3 2026, Saturday, 11am EST: Will be analyzing footage leading up to the J4 hand on Hustler Casino Live stream.
  • Jan 10 2026, Saturday, 11am EST: Analyzing the footage leading up to and including the J4 hand and dynamics
  • Jan 17 2026, Saturday, 11am EST: CANCELLED DUE TO SICKNESS
  • Jan 20 2026, Tuesday, 5pm EST: Talking about the J4 hand dynamics and cheating accusations, with Limon, host of PokerSesh
  • Jan 24 2026, Saturday 11am EST: Analysis of footage following the J4 hand
  • Jan 30 2026, Saturday 11am EST: Part two of analyzing the post-J4 hand footage

Live poker tells analysis, without hole cards: A first-of-its-kind stream (that I know of)

Starting in December of 2025, I (Zach Elwood) will be doing some occasional live-streamed poker tells analysis. The initial idea for this was to do a blind (hole-cards-hidden, commentary off, bias-free) analysis of streams I’d never seen before.

This was an idea I had after doing some similar analysis for a high stakes poker player I’ve been working with. I’ll say that this is kind of a risky endeavor, in terms of perception: doing this analysis “without a net,” in the open, will involve me making guesses and often being wrong, and that means there will be a good amount of footage where I look silly (especially if one compiled bad guesses). I think the riskiness to reputation is one reason no one’s done something like this before (that I know of). But I’m up for the challenge and think it will be fun; I myself have been wanting to do more of this on my own, to test my knowledge/reads, and I might as well do it in the open.

Also, I thought it would be fun to review the Robbi Jade Lew J4 hand and the footage leading up to that (I’ve talked about this for my podcast before but I’ve watched very little of the preceding footage). I thought doing the J4 hand might be a way to get some attention and draw some people in (thanks to Limon for the idea). If you have any ideas for interesting games/footage that people might like to see analyzed, email me your ideas).

Rules for submitting poker footage for analysis: 

  • Must be a cash game
  • Must be Texas Hold’em
  • Must be unedited (no compilations of exciting hands)
  • Zach will reject footage he’s analyzed before (scout’s honor on that, but fwiw I’ve watched very few poker streams the last ten years)

FAQs and details

  • Why am I getting so many things wrong? As I emphasize in my books and videos and other content, even players good at reading tells will likely only use tells to change a decision a few times a session, and sometimes (when playing tougher players) maybe not at all during a session. And behavioral reads can be useful in poker even when far from 100% reliable; this is because many decisions in poker can be close, so you don’t need much extra information to sway a decision. For this stream, I’m doing an experiment going in blind and testing reads. It’s a certainty I’ll frequently get things wrong; that is the nature of the project — but that’s also probably why no one has done this kind of thing before, as there is a risk of looking quite foolish. This is basically something I wanted to do on my own, to see how I did, but figured I’d do it live.
  • What are these confidence levels you’re using?  Related to the above points, a lot of the stuff I say on these streams will be highly speculative. That’s why I decided to try to use a 1-4 confidence rating when giving guesses. When I make a guess about hand strength, I’ll give a 1-4 rating, where 1 = basically a guess; not something I’d ever likely base a decision on, all the way up to 4, which means I’m quite sure that my read is correct.
  • Why are you so disorganized? If you watched the first stream on Dec 13, that was the first time I’ve done something like this, so it was a bit of an experiment. There are a lot of things to keep track of, and I’m working things out as I go. Appreciate the patience. This is also the first time (that I know of) that anyone’s done something like this, so there’s no pattern to follow.
  • Why are you getting the pre-flop action wrong so often? It can be hard to tell what happened early in a hand, due to the camera not catching bets/raises. Also, there can be straddles, which are hard to catch when the audio off. Also, sometimes blocking the hole cards also blocks gfx that help make sense of the action. I will mostly be focused on post-flop behavior, as pre-flop behavior is usually so ambiguous and high-variance. Put another way: I usually won’t care much what happens pre-flop, especially because I’m trying to focus on behavior and not someone’s strategic line in a hand.
  • Why are you ignoring chat? I will most likely ignore chat. There are so many things going on with this, even without chat, that it’s hard for me to focus, so I need to cut down on distractions.
  • Why are you speaking out loud about the hand details? I am using AI to summarize the transcript, so I want to include some details about timestamps and player names and such, for the notes.
  • How did you choose the stream? I’m choosing streams I’ve never seen before (you’ll have to take me at my word on that; but the amount of things I get wrong should help show that that’s true). I sometimes might take footage ideas from the audience and choose one at random.
  • Aren’t you biased by various factors, even with hole cards hidden and commentary off?  Yes, for sure. Just the way the hands go down will bias me: the hand history. For another thing, most large value bets will be strong hands; so that’s just a fundamental bias right there. And there are all sorts of other things that can bias me (e.g., player tightness or bluffiness). So my guesses will be biased in various ways; I’m just trying to remove the most obvious behavior-related sources of bias by blocking hole cards and related gfx and commentary.
  • How are you hiding the hole cards from yourself? I’m using a Chrome plugin called Hide Anything, which lets you drag blocking windows over any part of the screen. Then I’m capturing the video of my entire desktop and streaming that from OBS to Twitch.
  • What’s a major difficulty in this kind of analysis? One major difficulty in doing this kind of work is that poker footage often focuses on the player whose turn it is to act, and not the player who just bet. This means, for example, on the river, when a player makes a big bet, it often will cut away from him and we won’t get to see his behavior, and that’s one of the most useful times to get tells/reads on someone.
    • Another difficulty is that I have a lot to say about verbal tells (as I’ve written a whole book on it, Verbal Poker Tells). But unfortunately I don’t have an easy way to listen to verbal statements while also blocking the audio commentary.
  • How long will videos be available for? They’ll be up on the platform I recorded the session for a while (Twitch or YouTube live stream), but I’m not sure how long. I will likely also put them on my Reading Poker Tells YouTube channel