NoIQ, No Problem

[peterbirks]

So, NOIQ is leaving the IPoker Network. I was going to write that it had been chucked out, but that would probably generate a flurry of denials from NoIQ.

The IPNetwork statement read that:

"iPoker has, following discussions with Tain, decided to remove Carlos Poker and NoIQ Poker from the iPoker Network. The said card rooms will be removed from the iPoker network liquidity by January 24th, 2009."

NoIQ said that:

"This partnership is about to end January 24th. NoiQ and iPoker don't look at things the same way and it has been hard to understand each other."

It's no secret that NoIQ has a high proportion of the players that make the games on IPNetwork virtually unplayable, mainly because of its "VIP" programme (for which, read, "rakeback"). But the concern at IPNetwork probably isn't so much that NoIQ generates a bunch of very tight short-stack players, as that NoIQ consistently takes out more money from the network than it puts in.

You may recall how I wrote some months ago about the curious dichotomy set up by the "skin" poker system, one which does not seem to have been mentioned very loudly elsewhere. Put simply, a single poker site (e.g., Pokerstars) ideally wants all of its players to lose money slowly. It can't achieve this, because the skill of players follows a steady curve upwards. So, if you discourage the "winners", all you do is reduce your turnover and create a new stratum of "winners".

However, if you have a network with skins, then the network wants all of its skins to lose money slowly, while the skins want their players to win (at the expense of other skins). Why? Because the skins earn money by generating player hours. In effect, for a network, the likes of you and I aren't the players. It's the skins which are the players. The networks, as it were, are wholesale mortgage lenders, while the skins are the retail front-end.

Now, while Pokerstars can't improve things by "slicing off" a layer of players, in the network-skin-player relationship, things are (or, rather, appear to be) different. Firstly, the gradation of "skill" is less smooth. Secondly, the rule of "one rake-rate for all" is not set in stone. Because we are at the wholesale level, pricing is much more bespoke. I suspect that IPNetwork, roundly pissed off that it was paying NoIQ to take money out of the system, wanted to pay NoIQ less per player-hour, or rake generated. But everyone is being very tight-lipped. And let us not forget that Playtech, which powers the IPNetwork, is, I vaguely recall, now effectively a William Hill operation, and William Hill is a short-sighted bookmaker run by not-very-bright non-gamblers. If they took a look at the IP system and saw the NoIQ (and Carlos) numbers, they would probably take the immediate line of "close the account". I've no evidence whether William Hill was involved in the decision in any way -- but I do know what William Hill does to accounts that win regularly.

There's a plus side to all this -- the IP Network might, for a while at least, become playable again. However, the point that I mentioned about Pokerstars still applies to the wholesale end. By booting off the two biggest rakeback generators (probably to the cheers of all the other skins, who have been feeling right royally shafted by NoIQ) IPNetwork creates a gap in the market, which one or more of the currently cheering skins will be quick to fill. That's how capitalism works. So, in the long run, IPNetwork will just be reducing its turnover (quite drastically, I suspect, given the high number of player-hours generated by NoIQ) but won't create a "level playing field" where all the other skins lose money slowly. Another pair of "winning skins" will emerge.

___________

Comments

[(Anonymous)]

Now, while Pokerstars can't improve things by "slicing off" a layer of players....

This simply isn't true. The weak players will live longer - they drive the system. For any poker room there exists a detrimental player win-rate.

chaos

[peterbirks.livejournal.com]

Hi Chaos: I'm sorry, but this old chestnut has been debated again and again and, to my satisfaction at least, your line of thinking is flawed. There are two ways to show this. The first is theoretical and would take some time (certainly more than the character-limit in comment replies permits). The second is practical. That point is that no major standalone site has sliced off a top layer. Either they are all wrong and you are right, or vice-versa.

BTW, I don't disagree with your third sentence, just the first two.

I would also agree that errors can be made in creating games which are too big. But sites cannot gain by slicing off a top layer. Certain sectors of the player base can gain (because the "top layer" is barred, making that layer the big loser and all the other layers small gainers) but the site does not gain. And the overall "cake" is smaller. In a way, a slice off of the top layer is a bit like protectionism, and Adam Smith showed how that didn't work more than two centuries ago.

PJ

[(Anonymous)]

Hello Mr Birks,

I hope you're perkier now.

I'd very much like to see the analysis of any theoretical proof - so post/link away. I'll take your word for it that it's been done to death, I'm very occasional when it comes to following poker forums and now it's certainly been some time - I can see it being a 'death by 2+2' subject.

Anyway, you appear to assume that the new top-dogs replace the old top-dog's win rates. It would be trivial to produce a special case where this could not be the case (just look what happens if you get rid of the 7th guy called 'rake', of course it aint that simple); so I see no theoretcal proof. Where can the empiricial evidence be?

A few years back, I suggested that poker sites should, for their interests, educate players. The reason is obvious, and one you've alluded to, reduce the standard deviation in skill. If you cut off 5% you obviously do this. I can see an argument pervading that current higher echelons face slightly tougher opposition than those who would replace them once ejected (and also players pact stretch further apart once the very best go). And this would supposedly compensate for their skill short-comings against the prior elite; it may compensate, but sufficiently? You could construct scenarios where it were true, but so obviously where it isn't - where the skill-distibution tails aren't short and fat.

As for why cardrooms haven't done so, well, obviously, from a PR perspective it could be a disaster. Not to mention effectiveness - could you expect to keep determined players off the site? Too risky, one suspects on those axis. Besides, I'm surprised you've reasoned this way (that the evidence is there because none have acted so) - cardrooms are largely herd-like (insome respects, cartel like) in their behaviour. Just consider the move to high-limit cash games. It was a dumb move, but they all did, they all followed suit. Where we are today in-online poker is not the result of carefully planned or managed strategy, but a somewhat risk-averse stagger. If on-line poker gets tough, then the current model will be pressured - maybe initiating slicing off some winning players, cutting rake, reducing limits. But right now they aren't going to compete in ways that might fundamentally undermine their business, they don't need to: it's a gravy train. Only under the survival of the fittest, not the race to obesity, climate will you see the cardroom model optimised, I suggest.


As for fish driving the card-room well, I take your point, that the thing doesn't collapse if the weakest players disappear. Now admittedly, there is nothing you can do for the maniac, or the ultra-poor, but if you take care of the 'willing-weak', increase their longevity, then it will feed up and up and sustain the card room for longer. If you get rid of the cream, then you do this.

regards

chaos

[(Anonymous)]

'you appear to assume that the new top-dogs replace the old top-dog's win rates'

my bad.

[peterbirks.livejournal.com]

Except for the misplaced apostrophe, I'm not sure what you are worried about here!

As for the perkiness -- well, I've gone back to the old "Fuck you world, I'll show you". Such misanthropy is a useful survival mechanism and provides one with good financial motivation. But it's not recommended as a long-term solution, unless I want to turn into Gollum.

PJ

[peterbirks.livejournal.com]

I remember reading a v long thread on this on 2+2 about 18 months ago, and you will be pleased to know, chaos, that the majority appeared to back your viewpoint. However, they seemed to fall into your implicit assumption that:

"The reason is obvious, and one you've alluded to, reduce the standard deviation in skill. If you cut off 5% you obviously do this."

No, you don't, You only reduce the standard deviation if the 5% that you cut off are part of a non-bell-curve distribution. It's my contention (and this also seems to follow common sense) that skill levels are distributed amongst the total poker playing population in a fairly standard bell-curve. Slicing off the top 5% doesn't leave you with a bell-curve where the top 5% "cease to exist". There is a new "top 5%". Similarly, the "bottom 5%" gets slightly smaller and the top of the bell curve (the single hypothetical "average player") shifts slightly to the worse.

I accept your point that there is a herd-like mentality amongst poker rooms and also that it would be a PR disaster to make such a move. But I think that when it comes to hard cash, people are prepared to try to put good spin on bad PR, rather than suffer a loss of income. I really only chose this argument because I don't have time to go into a long argument on bell curves in a forum where I can't easily draw graphs! The "empirical" here is, I accept, a case of the elephant not being in the room.

I do feel that you see people as being intrinsically part of a single group (e.g., the "willing weak"). I don't see it that way. I see the "willing weak" as a variable feast. Cut out the top 10% and some of the "willing weak" become small winners. Some of the small winners become bigger winners and some of the substantial winners become the new elite. They aren't "moved" from one box to another, because the curve is gradual.

In a way, I think your viewpoint is Platonic ("a table has something intrinsic that makes it a table") whereas mine is a rather fuzzier viewpoint (where categories merge in grey boundaries). We automatically categorize things (tables remain "tables", even when people can sit on them, but then they become "benches", even when people use them as tables).

PJ

[(Anonymous)]

Pleased with a majority view point? Tush, tush. But my stats knowledge is pretty superficial, I'm sure there are some in-depth responses there, but I'd think modelling it would give the best clue.

No, you don't, You only reduce the standard deviation if the 5% that you cut off are part of a non-bell-curve distribution.

I addressed this somewhat implicitly (and clumsily) here:

(and also players pact stretch further apart once the very best go).

Or maybe not. If all the 7 ft 6 + people in the world were killed off you still wind up weith a bell-like curve, with a reduced standard deviation. Of course, in the annexed poker-player case, a bell-curve still exists, but it's a different one, it won't have as long and thin a tail. The best won't get as many exam-questions right as the previous lot - they'll miss opportunities the previous bunch wouldn't have, perhaps a few more might spring up, but they won't do as well: in addition the best drive the standard forwards, so the new wave might actually drop off (in quality) a little. I don't know why you assume the new bell-curve will replicate the old one, if it doesn't then we're open for business.

If say, the critical winrate was 3 bbs/100 and the site killed off all 2.5+, it isn't a given the 3.0 barrier will be breached again (it might). You are claiming their shoes will be fully filled by the next wave: it won't. It would be too spooky if they did. So now we'd be talking trade-off. Their win rates will improve becasue the opposition is now weaker, but there is no basis to assume they will reach the same lofty heights and damage the card room as the previous bunch did.

The term was 'willing weak' was ambiguous at best, or just outright misleading. By 'willing' I meant good-spirited, so 'win if they could', not 'willingly weak'. A maniac, of course, is a maniac, is a maniac.

There are, of course, soft issues here too, the very, very good can be game breakers, there are loss rates, or simply playing environments, that cause some weak-players to capitualte.

Aside: I'm anticipating a fiscal 2009 predictions, tips post.


chaos

[(Anonymous)]

'Certain sectors of the player base can gain (because the "top layer" is barred, making that layer the big loser and all the other layers small gainers) but the site does not gain.'

I didn't explicitly address this point. While the site may not gain post-eviction from those that were winners when 'top layer' was in play or indeed from those with deep pockets, short time, they are in the minority. The poor players do get an increase in exposure, but marginal, but those near the margins - just winning/losing will likely witness huge increases in game-time. Previously, they might have been restricted by funds, now they are limited by time.

[codebonuspoker.livejournal.com]

Ok you have some defined and real ideas on this just one thing im missing, skins and networks make money just by people playing at the site, so in your opinion was NoiqPoker making a negative move by pushing the rewards and by doing so turning normal players into serious rakers and thus where they "shocking" other skins? I mean was that great impact that a lot of cash outs were made by their players, when i know a lot of them, and they actually played the bigger games in the room and the jackpot sng and i can tell you some might get some money back and get occasional 50K jackpot, but still in all those skins you really think that was the issue here? guys playing 50+9 sng tabling 10 of them sometimes and making 90$ rake in each run... plus all those 6 table Kamikazes, this is serious money for the network. Imagine a big player base like noiqpoker. I think most of this was just related with how the cash \ vip programs were sold in affiliates sites.... i think entraction will be good for the noiqpoker concept, half a million rake race for starters...

[peterbirks.livejournal.com]

A neat line in political sohistry here?

skins and networks make money just by people playing at the site, so in your opinion was NoiqPoker making a negative move by pushing the rewards and by doing so turning normal players into serious rakers and thus where they "shocking" other skins?

In other words, "Networks make money by having lots of players playing a lot. NoIQ encouraged players to play a lot. Therefore NoIQ was good for the network."

See what you can manage with a bit of spin?

As for the half a million rake race. Well, fuck that. That's allocating money to groups whom I could never hope to join -- multiple players on single accounts; youngsters with lots of stamina, no boredom threshold and plenty of time on their hands. Rake races are awful things for me, because they actively encourage multi-table short-stacking to a rigid formula. I can beat most of those players, but, hell, just look at the masses of "8% VPIP" averages caused by these players.

I think this has hurt NoIQ badly and the Entraction thing is the best spin that they could put on it. They would far rather have stayed within the IPoker system. On the plus side, this will enable the objective observer to see what impact this has on (a) player numbers on IP Network/Entraction and (b) tightness/looseness of games. That should enable us to see roughly what proportion of the IPNetwork was made up of NoIQ players and (to a lesser extent) what proportion of those players were making the poker games unplayable (from anything but a mass multi-table rake-accumulating viewpoint).

PJ

[codebonuspoker.livejournal.com]

yes, you are write we now shall see if it was really noiqpoker players the one that played short stacks in a system... i dont think so but i cant be sure of course i have the idea a lot of them were very well rolled and had no need to play half a buy-in. I dont play limits high enough to confirm this nor i know every nick out there. On what the rake races is concerned, well the average fish on Entraction Network WILL suffer in the coming months :D that's for sure, i imagine that some quality player in other skins will manage to get an edge from all that action. I really cant predict but i will keep posting here with any feedback i have from my players. Both the ones moving to Entraction and the ones moving to other iPoker skins will give me some first hand impressions. I will keep you posted.

Warm regards