Author Archives: Noah Stephens-Davidowitz - Page 2

Absolute Poker Keno Update: Money Repaid. Explanation Still a Lie and Reveals Further Amazing Incompetence and Sketchiness

Eight days ago, I wrote a post about Absolute Poker’s ridiculously non-random Keno, which detailed a pathetically incompetent mistake that they had made (or perhaps that an outside contractor, Betsoft Gaming, had made that they’d completely failed to notice). It also explained that their official explanation was a lie and that over five months had gone by without compensation or a better explanation. (I highly suggest reading that post before this one. Otherwise, you’ll have absolutely no clue what I’m talking about. Plus, it’s worth the read.)

Well, I’ve been paid back. At 2:00 today, I got this e-mail from AP (I bolded the important part):

Read more »

Life as a HU SnG Pro by the Numbers (It’s Awesome… If You’re Good and Have Rakeback)

In previous posts, I looked at how variance affects players who play large-field MTTssmaller-field MTTs, NLHE 6-max cash, and 6-man and 9-man STTs, and I found quite a bit of difference. Next up: HU SnGs. (Still to come: HU NLHE, FR NLHE, HU PLO, 6-max PLO, HU LHE, 6-max LHE, and FR LHE. I’m going to be quite busy for at least the next couple weeks, so please don’t hold your breath. Follow me on twitter if you’d like to know when these posts go up.)

HU SnGs are quite simple statistically, so I will make exactly one assumption: rake that is 1/22 of the rake-free buy-in. This is the standard rake for turbos on Stars and FTP at most stakes (e.g. $110+$5), so everything that I say will be exact for those games, but if you play games with a different rake:buy-in ratio, the numbers will be a bit different. To get most of the results, I won’t even bother to assume a normal distribution; I’ll just use the binomial distribution, which is an exact statistical representation of a HU SnG. (This won’t actually change the numbers at all after rounding, but it just requires a bit of extra algebra from me, and it’ll appease some of the statistical purists in the audience.)

Anyway, with that out of the way, let’s jump in. Say you’re a HU SnG pro with a solid ROI of 3% at the Stars $115s. You’re also moonlight as the world’s greatest physicist, and you invent a cloning machine that you planned to use to resurrect Einstein, Lincoln, Ghandi, etc., thus ushering in a new era of global age of peace and prosperity. But, then you remember that rent’s due soon, so instead you make 99,999 clones of yourself and get to grinding. You figure 1k HU SnGs each should cover rent plus some standard expenses. (Trips to the Rhino get expensive when there are 100,000 of you…)

How does this HU clone army fare? Good question!

Read more »

Absolute Poker Rigged Keno: Five Months Later with No Compensation and a False Explanation, and How This Relates to Superusers and Joe Sebok

(Update 4/1: AP has repaid customers, but their new explanation leaves a lot to be desired. I recommend reading this post first if you haven’t read it yet, but then see this post for the update.)

Cereus is in the news again, as UB sponsored pro and tweeter extraordinaire Joe Sebok has finally made a 2p2 account to talk about various things. Frankly, reading those threads is just about the most frustrating possible use of one’s time, but for the masochists in the audience, please accept my flower of links: (((1 2 3 4))). ($5 on Stars/FTP to the first person who correctly identifies that reference.)

Basically, what’s going on currently is an argument between Joe and 2p2 in which Joe insists that current Cereus management is clean and 2p2 argues otherwise. (Much of it might actually come down to Joe’s rather lax definition of cleanliness, actually.) Needless to say, the UB/AP superuser scandal is an incredibly big deal. But, it’s so painfully nuanced, complicated, and shrouded in mystery that answering a simple question like “Is Cereus currently run by a bunch of crooks?” is amazingly difficult. So, I’m going to leave the larger scandal to the professionals and sidestep the issue entirely to discuss a much much smaller on-going scandal: Absolute Poker’s rigged keno game and their response. I think that that scandal deserves some more publicity in its own right (and it’s entirely my fault that it has not gotten enough), but I also think it should provide some perspective on the current discussion.

(I think it’s worth noting here that, though this scandal is several orders of magnitude smaller than the superuser scandal, had it happened on any other US-facing network, it would have been huge. The fact that it’s received such disproportionately small attention from the poker community (myself included) is a testament to how jaded we all are when it comes to Cereus.)
Read more »

Life as an STT Pro by the Numbers (It’s a Lot Better Than You Probably Think)

In previous posts, I looked at how variance affects players who play large-field MTTs, smaller-field MTTs, and NLHE 6-max cash. Now, I thought I’d grab some low-hanging fruit in the form of sit-n-gos. It turns out that 9-handed and 6-handed STTs are very similar statistically, so I’ll lump them together below (I justify this in my assumptions section). HU SnGs are next in line, and should be done in a day or two (no promises).

If you’re a fellow nerd, you might want to read about [slider title=”my assumptions”]

  1. I’m only going to consider Poker Stars $114 9-mans and $119 6-mans. Some sites have different payout structures. In particular, some sites spread 10-mans and/or 5-mans instead, which obviously changes the payout structure and changes the numbers as well. This analysis will still give a decent picture for all roughly similar games, but keep in mind that it is explicitly an analysis of the Stars $114s and $119s.
  2. I’m going to assume normality. STTs are pretty close to normal over samples of 100+ tourneys and essentially indistinguishable over 500+ tourneys, so that shouldn’t be a problem. This follows directly from properties of the binomial distribution.
  3. I’m going to assume constant standard deviations. In theory, standard deviation for an STT player is dependent on her win distribution. So, players with different ROIs can be expected to have different standard deviations, and even players with the same ROI could have different standard deviations. In practice, these effects are tiny: Standard deviations vary by only about 20% in 9-man STTs and only about 10% in 6-max STTs over reasonable finish distributions for serious players. I’m not looking to estimate confidence intervals within 10 or 20%, so this should be fine.
  4. I’m going lump 9-man and 6-man SnGs together, with a 1.5 BIs/tourney standard deviation for both 6-max and 9-man STTs. I didn’t initially plan on doing this, but it turns out that the numbers are almost identical for the two. Typical standard deviations are about 0.05 BIs/tourney higher for 9-man STTs and about 0.05 BIs/tourney lower for 6-man. So obviously this approximation is good enough for my purposes.

(Of course, if you know basic statistics, everything in this post is derivable easily from the above. So, the real meat of this post is contained in the assumptions, which are all justified by a bit of behind-the-scenes research with sharkscope and windows calculator and some discussions with friends of mine. The rest is essentially just watching me divide by SQRT(n) and plug in to my favorite z-score calculator repeatedly)[/slider].

In my previous posts, I didn’t consider rakeback because it typically varies by stake and becomes much less important at the higher stakes. STT players make a large percentage of their income from rakeback, VIP programs, and bonuses, even at the highest stakes, and it doesn’t vary much by stakes. So in this post, I’m going to talk about “effective ROI”, not ROI. Effective ROI is a phrase (that I made up) that means your ROI after you consider rakeback, bonuses, VIP rewards, etc. In other words,

[latex] \displaystyle (\mathrm{Effective\ ROI}) = (\mathrm{Raw\ ROI}) + \frac{(\mathrm{Rakeback\ etc.}) }{(\mathrm{Buyins,\ including\ all\ rake)}} [/latex]

 

With that housekeeping out of the way, let’s look at the numbers. Take a player with an effective ROI of 7%. (He might, for example, play $114s with a raw ROI of 4% and earn an additional 3% from Stars VIP program and bonuses–equivalent to a RB % of about 38%.) What happens if he and his 99,999 identical grinder twins (!) play 1,000 STTs each? Well… this does:

Read more »

Request for Data

I’ve got more stats posts in the pipeline (i.e. more of these: 1 2 3). I don’t need any extra data for any form of STT (9-max, 6-max, HU), so I think I’ll either do 9-man SnGs or some bankroll stuff next.

However, I do need some additional data for the following games: NLHE FR (50 BB+), PLO HU (50 BB+), PLO 6-max (50 BB+), LHE HU, and LHE 6-max. (I can use my own data for NLHE HU.) If you’re a professional with a lot of hands played at any of those games, please get in contact with me. You can PM me on 2p2 or deucescracked or e-mail me at noahsd (at) gmail (dot) com. Please mention what games and stakes you play, what site(s) you play on, how many of your own hands you have in HM (you must have HM), and your winrate.

You won’t actually be sending me your hand histories, so you shouldn’t have any ethical or privacy concerns. (I’ll just give you instructions to dump the data that I need to a spreadsheet.) You can remain anonymous if you’d like, or I can give you a shout out if that’s your thing.

If you’d like to see me analyze any other type of games (PLO FR, various forms of CAP, various forms of stud, O8, etc.), let me know. Keep in mind that I can’t do anything with games that aren’t supported by any tracking programs.

Life as a NLHE 6-max Cash Game Pro by the Numbers (It Ain’t Too Bad)

(Update 3/8: I originally completely forgot to talk about what happens with different winrates. I added this analysis to the bottom of the post.)

In two of my previous posts (1 2), I crunched some numbers to show that the variance in large-field MTTs is pretty damn crazy. Now I’m going to turn my attention to NLHE 6-max cash games. (Sorry for the delay. I could claim to have been busy, but mostly I’ve just been lazy.)

This is actually pretty easy thanks to the statistician’s best friend, the central limit theorem. For sample sizes of at least a few thousand hands, you can just take your standard deviation and winrate and use the normal approximation. (For sample sizes of less than a few thousand hands…. Who cares?) This makes the math really easy, and in a way, it’s responsible for the relatively cushy lifestyle of cash players–The normal distribution is a lot cleaner than the distribution that I found for MTT players. (Proof that the distribution is normal)

(Another consequence of this mathematical convenience is the fact that some other people have done this analysis already. While my posts on MTTs were, as far as I know, the first honest attempt at tackling that problem correctly, this post will mostly just explain what’s already known to the nerdy contingent of the poker world and anyone else with a basic understanding of statistics. I’m just bothering to share this information with our less nerdy brethren in a way that I feel is reasonably clear. To that end, I’ll mostly just leave out any explanations, but suffice it to say that all of this comes from very basic facts about the normal distribution. However, I think everybody who plays poker should be able to do some basic statistics, so I might make a tutorial explaining where these numbers come from at some point.)

To see how that works out in practice, let’s look at a basic example. Take a solid 6-max NLHE grinder with a 5 bb/100 win rate (I’m going to use bb/100, not PTBB/100 in this post. An unfortunate tradition ported from LHE leads many people–including PTR, PT3, and sometimes me–to call a BB or a PTBB twice a bb.) with a fairly typical standard deviation of 90 bb/100. What happens if 10,000 clones of this guy play 50,000 hands each? Well, this does:

Read more »

Your Body Is Awesome and Really Complicated; Please Don’t Trust It To Idiots and Charlatans. (No Poker Content)

(Seriously.. No poker content at all.)

The human body is unquestionably the most complicated machine that any person has ever encountered, and it’s likely to hold that title. Each of the about 6.8 billion people on Earth is composed of a unique set of roughly 100 trillion cells (Whenever you hear the word trillion, be amazed) of a huge variety of types. Each of these cells is way more complicated and elegant than your high school bio textbook said. The organelles in your cell exist in this terrifyingly confusing and surreal world in which incredibly weird objects of a huge variety of scales interact according to the bizarre laws of organic chemistry, three-dimensional geometry, and quantum mechanics. This is explained way better than I ever could by this insanely awesome video created for Harvard students:

Read more »

Life as an Online MTT Pro II: The Numbers Are Back, and They’re Out for Blood

(This is a follow-up to this post. You probably want to read that first if you haven’t already.)

My previous post about the variance in MTTs got some interest and plenty of criticism, so I thought I’d follow up with a (slightly) less sloppy post that considered some criticism. In future posts, I plan to look at various forms of cash games, STTs, and DoNs. (If you’d like to share your data for any of these things, please get in contact with me.)

First of all, I should clarify something: Some people are definitely better off playing large-field MTTs than other games. I’ll make an argument below that almost no small/mid-stakes players fit into this category in the current climate. For mid/high-stakes player: If you’re insanely good at MTTs and you’re down to put in lots of volume, then play them. If you really enjoy large-field MTTs and you’re willing and able to deal with their ridiculous variance, then play them. If you’ve got a really sweet backing deal, then play them. If for whatever other reason you think you’d like to play them, I won’t pretend to know what’s best for you. But, make sure you seriously consider the extreme variance that’s involved. Before this, I haven’t seen anyone present this information correctly. Thus these posts. (I have seen people use the normal approximation to look into this, but as I showed here, that method doesn’t actually work.)

Anyway, there’s lots of fun stuff that I didn’t do with Shaun’s data that I’d like to do with my shiny new data set (see the section at the bottom for more about this sexy piece of data). For example, varying buy-in sizes obviously increases variance. For Shaun’s data, I looked at buy-ins between $55 and $216, i.e. with a max buy-in about 4x his min buy-in. But what if we don’t vary the buy-ins at all? Here’s the data for that:

Read more »

Life as an Online MTT Pro by the Numbers (It’s Hard)

(After you read this post, you might want to check out my follow-up to it here.)

I tell a lot of people not to play large-field online MTTs for a living. I’ve always thought that the variance is just way too high for most professionals to trust their livelihood (and sanity) to large-field MTTs instead of cash, smaller-field MTTs, or STTs. But, admittedly, I’ve given this advice without any direct evidence to back it up. I’ve been meaning for a while to see what the numbers say, and this post will be a tentative first step.

Ideally, what I’d like to do is do a nice controlled study where I pick a few representative players based on past results and use their results over the next few months as my data. (Alternatively, I could take the results of one of the large backing groups. If anyone who backs 20+ people would be down to share some information, let me know.) But, that requires more motivation than I’ve been able to muster, so I decided to do a much rougher study: I grabbed Shaun “SFD” Deeb’s tourney results from OPR (with Shaun’s permission) and played around for a few hours. Here’s what I found:

Read more »

Unenforced Rules Suck

It’s becoming more and more clear that the major poker sites are not enforcing many of their own rules.

Part of this is because they’ve made rules that simply can’t be enforced: FTP bans ghosting, and most major sites now ban datamining. Some of their rules clearly are enforceable but either aren’t enforced at all or have no real punishments associated with them: I only know of one example of a site actually confiscating money for multiaccounting in cash games (and it was a complicated case), and I don’t think anyone’s ever gotten more than a warning for using PTR while playing. (Similar problems exist in the live poker world as well, but I’m not really qualified to comment. Nate detailed a bunch of problems with selective enforcement at the PCA in this 2p2 post.)

The result is, predictably, a lot of confusion. Some people simply ignore all these rules and make a lot of money as a result. Most people ignore some of the unenforced rules (PTR, ghosting while coaching), but not all of them. Some people get in trouble for doing things that they didn’t even know were wrong. Throughout this process, the unquestionably important rules such as the bans on collusion or buying accounts deep in tournaments lose their weight. This is obviously a terrible situation, and it will only get worse if the sites don’t do something about it as people continue to learn what they can get away with. Current high profile cases of people breaking the rules and making tons of money off of it with no consequences, like ugotabanana and PTR have, will encourage others to follow suit and certainly won’t make things easier.

So, things definitely have to change. In each case in which a rule is either not enforced or enforced only selectively, each site should either change the rule or start seriously enforcing it. They need to make it clear that breaking their rules is cheating, and cheating is both unprofitable and unacceptable. I’ll outline my specific ideas on how to handle multiaccounting and datamining below (a lot of which is just copied and pasted from an old 2p2 post of mine), but I think that the general policy that rules are rules is much more important than the specifics.

Read more »