bwin’s New Rake Calculation

bwin recently sent out this email to affiliates:

Dear Affiliate Partner!

bwin is pleased to inform you that a new algorithm called “Essence” will be introduced on Thursday, 15.07.2010. This algorithm will finally introduce a system where the player’s overall contribution to the poker ecology becomes a major component to its healthy balance.

From this date onwards, Essence will take the following issues into account:

the player’s style of play, pot contribution and skill level
the long-term effects different players have on each other
The purpose of this new procedure is to evaluate a player’s true value – to assign the correct rake amount – and to ensure long-term profitability for both you and bwin.

All players are valuable within the poker ecology, hence bwin will not use any form of penalties or sanctions to ensure a healthy balance. By using Essence to assign rake on an individual player level not purely based on activity, we will ensure that there are incentives to attract all types of player, and that no players will be unprofitable.

Most Affiliates will notice an impact on their earnings as a direct consequence of the redistribution of the rake.

Please note that our Poker Points system for bwin users will not change. The points awarded for cash games will remain at 3.25 Poker Points per dollar in rake and tournament Poker Points will stay the same.

If you have any questions, please do not hesitate to contact your Account Manager.

Your bwin Affiliate team

This is a rather interesting development because unlike iPoker’s response to skins that produce too many sharks bwin has opted to adjust the amount of rake that they’re willing to pay affiliates based on various factors mentioned in the email. But this really only works if you get the algorithm correct.

That is going to be the devil in the details because I’ve seen plenty of people try to come up with a way to calculate a player’s net contribution to the poker ecosystem but none have done so with enough accuracy that I’ve ever heard or seen anybody actually use that metric.

While I was writing this post I received a second email from Betsson (on the Ongame network) that adds some more detail on exactly how they intend on arriving at this magic number.

In Essence each player is assigned a coefficient, based on the past 90 days’ performance in terms of the monetary end result. The coefficient is transformed and normalized to reflect the relative skill gap between the players in the hand. Until they have played 500 hands, players are assigned the default coefficient.

Each player is assigned an amount of rake based upon their coefficient in relation to the opposition’s coefficients and the amount of money they contributed to the pot.

If all players at a table have the same skill level, then the rake would be distributed in the same way as before Essence.

Essence will only affect cash game rake. Tournament rake will still be collected as previously; the house-fee set on tournament buy-ins.

Loyalty program/Player points – Players will see the difference between the old and new setup in the way they accumulate player points in the loyalty programs, players will receive more or less points than before depending on their style of play.

Fully integrated real time solution – Since the model is a change in the core rake algorithm it will be fully integrated with all systems cascading the player re-evaluation throughout the value chain in a transparent way.

Essence will make it more valuable for affiliates to send casual players to Ongame Network rather than grinders and pros.

Yes, they’ve added more detail but they still don’t tell you how they arrive at the coefficient or exactly how they go about normalizing.

While I think bwin is going in the right direction it does concern me that they don’t publish this formula. First, players have a right to know how their play is being valued. Part of any winning player’s ROI calculation involves the monetary and non-monetary benefits of the loyalty system in addition to their win at the tables. Now there is no clear way to know how many loyalty points are going to be earned.

My second concern is that unless bwin/Ongame are sharing this algorithm with the skins this could quite possibly turn into a massive blunder. If I was a room on the Ongame network and I can’t figure out how to encourage players to play optimally on the site in order to maximize my return on marketing dollars then the entire platform losses a huge amount of value.

I think we can all safely assume that total fish who keep depositing and losing money will result in maximum value and that grinders will yield the worst value. But when does the fish start to lose value? How much can he win before I (and the player) take a haircut on his play? Could I craft promotions that might loosen up some of my grinders? How much do I need to loosen them up? How far does their win rate have to fall?

Again, this is all a black box solution unless bwin publishes the algorithms that help them determine how many loyalty points are being assigned per each hand. And my third and most serious concern is that does bwin have the algorithms right?

No offense to the good folks over at bwin and Ongame but previous results don’t exactly build confidence that they can do what no other poker company seems to be able to pull off in creating a comprehensive player modeling algorithm that can account for different player types.

If they don’t get the algorithms right they could inflict huge penalties on certain classes of player unintentionally which will force them to leave the network altogether. Or perhaps knowledge of such a system causes players to experiment around with ways to optimize their winnings on Ongame so maybe knowing that it’s a rolling 90 day snapshot they quit playing after running too hot in order avoid penalties. Then they come back after 90 days with a fresh start.

Like I said, I think that looking at player behavior and valuing players differently is the right direction to go, but I’m just not confident that a closed-box system is in anyone’s best interest. It’s not even in bwin’s best interest. It will only raise suspicions. People will talk about it on 2+2 and other forums and decide to play where mysterious coefficients aren’t being applied making it impossible for them to forecast their ROI.

Some might say that’s sort of what bwin is hoping for. The site will be less attractive to grinders and will make it more profitable to bring in the fish. But the rake the grinders generate helps fund the marketing campaigns that bring in the fish. And to be quite honest, the vast majority of skins don’t know how to market to fish. They’ve built their businesses either cross selling their casino and sports books and they’ve already drained that pond or they’re so poorly underfunded that they cannot possibly run at a loss long enough to entirely switch around their business model from catering to a small group of grinders to becoming a marketing machine generating tons of first time depositors (the vast majority of which will never make a second deposit which is why this business model is so expensive).

The last few paragraphs in the email from Betsson to affiliates says quite a bit.

The maximum rake extracted over any period of time equals the losses of the losing players.

Industry mistake – all players have been valued in the same way on game activity no matter if they are winning money from other players or contributing money to the system by losing.

Each dollar won or lost by players has a financial value
There is a gap between current and real valuation of players
The gap has existed in the industry since day one
All players will be profitable even if value change
No negative coefficients

This looks like it was cut and paste out of a bullet points memo. It reeks of PowerPoint 🙂

I’m not quite sure I follow or agree with them on the “all players will be profitable even if value change” or “no negative coefficients” though those may have been out of context as a sub-point to “the gap has existed in the industry since day one” but got lost in the formatting copying and pasting it from the bwin/Ongame presentation.

Anyway, they’re right that there is a gap between current and real valuations and that gap has existed since the industry began. That’s exactly why I think other rooms looking at winning players as leeches on the system is so wrongheaded. All players bring value. Even the apex predators in real life ecosystems have a place in the food chain and the extinction of one level in the chain disrupts all of the other levels.

That being said, I sure hope bwin/Ongame got this right because if they didn’t the fallout could be pretty huge. If I was any of their competitors I would be marketing like hell to defectors.

5 thoughts on “bwin’s New Rake Calculation”

  1. I don’t think they need to know it WILL work, but they do need to know it might work. People take a shot all the time not knowing whether it will work or not. Knowing that your rake will be low/non-existent while you start out should go a long way to increasing people taking a shot.

    Some will succeed and be happier/wealthier than they were before (Group 1). Others will drop back down and continue to grind at the lower level, which should be +EV since the best of the best will have moved upstream (Group 2). The third group (Group 3) will bail on the site and move to a competitor (which should also help the Group 2 by making the level easier). The risk is that Group 3 will be too large, but look what you’ve also done. Group 4 (losers at the current level) who are likely to stop playing, should be encouraged to continue since they are being financially rewarded by less/no rake. And Group 5 (the players at the level above) are being helped because of the increased players taking a shot are adding money upstream.

    You are correct that folks need to know how this affects them, though. They don’t even need to know the specifics of the calculation, but they need to know ahead of time what their rake rate will be and be notified ahead of time of any changes to their status.

  2. @Chris: But only if they know that moving up will work. But if nobody knows how the model works you may be moving up and getting hammered at the same time.

    @Marco: Exactly! By not knowing until after the fact it makes it difficult for me to determine whether or not eating at this restaurant is a good value. It’s only when they bring the check over for $500 for two appetizers that I find out 🙂

  3. Hopefully they do it better than that. I would hope they would establish levels and notify you of a change in level before they started raking at the new level.

    The messaging would be more like ‘its like the Income Tax. Different people pay different percentages of income tax, based on their income.’ That is something the players can relate to, even if they don’t like it. Then, for those who need/want more convincing you can (hopefully) show them that grinders are better off either a) moving up in stakes or b) staying where they are and having lesser competition because so many of the other grinders have moved up that the competition is softer.

  4. BWIN has been talking for a long time about the value of customers. So finally they took action.
    What I don’t understand is how they want to communicate it to the players. It’s like if you are dining in a restaurant and the owner telling you: I will decide on the price you have to pay for the meal based on the tips the other guests payed during the past 90 days, but I cannot give you an indication on the price before you have ordered the meal and requested the bill.

  5. What if we think about this, not as a grinder penalty, but as an up-stakes inducer? If done correctly, the winning grinders could be encouraged to move up in stakes, where they’ll become a beneficiary of the system by being at the bottom of the rankings at the new stakes. It makes ‘Taking a Shot’ more manageable. Grinders look at the bottom line and moving up in stakes successfully will often be more profitable than they were before.

    I think its a great idea.

Comments are closed.