Inventing PokerStars

Why this and not that...

Inventing Pokerstars"When you come to a fork in the road, take it." -- Yogi Berra

I became involved with PokerStars ten months before the site launched for real money. Almost immediately, the first of many choices (forks in the road) presented itself. As the project developed, often two or more paths diverged into the unknown.

One very early example was the question of game structure. The decision could plausibly sink the site right from the start, but this was one answer I was certain of.

In the brick & mortar world there were two competing ways of doing things. In California, games were usually 9-handed, with a bet and three raises in limit games (all games were limit back then). In Nevada, 10-handed games were common, and a bet and four raises was standard. Paradise Poker, the 90% online industry leader at the time, offered a hybrid -- 10-handed games with a bet and three raises.

As a California player, my view was 9-handed with three raises were the clearly better choices -- if our priority was entertaining, action games where the pot easily surpasses the rake cap most hands. Back then, California games were very live with multiway action, while Nevada games were far tighter, often ending up with head-up action. So, I wanted PokerStars to try to always use California rules.

At the time, our choice to go with 9-handed and 6-handed games (in contrast to Paradise's 10- and 5-handed ones) got complaints from very tight rock players. I wrote then "Nine handed games have more action and hurt the rocks" but entertain recreational players and afford active winning players a good environment to battle in. I'm kinda proud of myself for recognizing that then, but ticked off at myself for not recognizing the same issue when it came to the establishment of the rewards program that unfortunately rewarded rock players!

Despite going against the Paradise standard, time has demonstrated this was the right choice. The vast majority of players prefer 9-handed to 10 (and of course many prefer even going further to 6-handed games or head-up).

How NOT to test tournament software
Prior to PokerStars launching, the scam site Pokerspot offered the only multi-table tournaments... but these had a truly bizarre flaw. If three tables were left, you could have two tables of nine players and one table of four. They had no means of balancing tables. One of the first things I was asked was: "what do casinos do in this situation?"

Right then it hit me, no one else in the company had any meaningful amount of experience with brick & mortar poker. Most of the people involved were software people. Besides myself, the only other "poker people" were Terrence Chan, a 20-year-old online poker player Isai hired because of his postings on the Rec.Gambling.Poker newsgroup and John Harkness, a movie critic who played poker as a hobby and also posted on RGP. Terrence went on to tackle the Herculean task of setting up the PokerStars support infrastructure, but initially his practical experience was limited. John mostly just offered the perspective of a recreational player. I quickly faced the reality that it would always be up to me to answer the question of: "what do casinos do in this situation?"

After explaining the different ways casinos balanced tables at the time, I suggested taking the player from the closest to the big blind position at a fuller table and moving that player to the closest to the big blind position at the shorter table. Since no one had any reason to disagree, we went with that... so table-balancing would be one of the features to really pay attention to during the first No Limit Hold'em pre-alpha multi-table tournament I played in.

Since there were only a dozen humans who could play, the tournament field was filled out by 100 "dumbots". These dumbots randomly performed fold/bet/raise, and they only knew two betting sizes: the minimum or all-in. So before we start, in the master chat box we go over the features to pay attention to, including how the software handled table balancing.

With humans spread at different tables, the tournament began... and literally fifteen seconds later John Harkness typed in the chat box that he was eliminated the first hand, when he tried to bluff a dumbot! We all began teasing John, saying that we were hoping to test the check and checkraise buttons too, not just the bet-and-be-eliminated functionality.

John's sacrifice was not in vain though. After it became clear the table balancing functionality was going fine, I took up the ultimate challenge for a professional poker player... pad my player resume by beating out one 20-year-old Canuckian future mixed martial arts fighter, nine software developers, and 100 dumbots.

Silly as that idea started out, it actually became a challenge to beat the army of dumbots. Since the dumbots had no sense of hand values or pot equity, the idea was to get them to call a largish bet preflop... then bluff them the minimum bet on the turn and river. This especially was the plan of attack when I was in position. When first to act, a dumbots choices were check or bet... but checking to a dumbot meant check-folding. This meant that 50% of the time a dumbot would call a preflop raise they would check-fold the flop, and half the time they would bet, they would bet the minimum.

Anyway, I got down to headup against a dumbot that had me outchipped 5-1, and the bot busted me when I flopped a straight and the dumbot turned a flush.

The first public tournament
In late August the dumbots were erased and we were ready for our first alpha test of the software. Now besides the dozen people involved, I invited Shirley and a half-dozen other friends to be the first outsiders to play on the software... and to try and find ways to make it fail that we hadn't thought of. Everything went smoothly, so the next step was a higher profile, invitation-only tournament (#295) with some small cash prizes ($200/$100/$50).

We compiled a list of about fifty influential industry people, active RGP posters and other people I knew. One remarkable thing about that first tournament is that of those 50 players about a dozen ended up working for PokerStars in one way or another. The tournament had some glitches but went okay. Sharon Goldman won the invitational, despite laboring under the burden of playing under husband Dan's name. Even though it was well-known I had contacts at most of the online cardrooms, Dan was one of the less than a half dozen people to ever ask me about getting a job at online cardroom... which lead to him being hired as PokerStars' first Director of Marketing.

Of the 50 invitees, one considered my email invitation spam, and one, Mason Malmuth, declined the invitation by writing he found it "amazing" (not in a good way) that I would send him an invitation to anything. This "amazing" response became a running joke over the years since I was constantly advocating spending millions of dollars each year on online advertising, especially on 2+2, including a proposal (see below) to spend $10,000,000 on 2+2 alone: "How pissed do you think Mason would be if he knew you are advocating spending an additional ten million on 2+2?" (As I mention here, at the height of the boom Two Plus Two, Cardplayer and the World Poker Tour missed out on tens of millions of dollars because they didn't recognize early on that "a rising tide lifts all boats"... meaning it was more important to work together to make the overall boom bigger rather than fight to dominate a smaller market.)

How NOT to launch an online poker business
The invitational tournament above took place the first week of September 2001. After tweaking a few things, it was decided we would launch the public beta, play money version of PokerStars the next Tuesday. (All online cardrooms had a beta period up to that point; for example, UltimateBet had a six month beta period.) The plan was for Terrence Chan to post a notice to the RGP newsgroup Tuesday morning. Terrence "wrote up this announcement, saved it on my computer, and went to bed." He woke up the next morning and hit 'send'.

That Tuesday morning was September 11, 2001. Only after hitting send did Terrence check out the news of the day to discover the World Trade Center terrorist attack was underway. Not only did we look like clueless douches for launching an online poker site during a terrorist attack, it was tremendously disappointing -- all our planning went down the crapper. I'm convinced the site would have done much better initially if we had launched the day before or a few days later. But in the end, PokerStars did become the overwhelming online poker leader, even if it did launch on the single worse day to launch a business this century!

Suppose your play money games are TOO good
PokerStars' beta period turned out to be quite successful, largely because we had multitable tournaments and no one else did. Even though they were play money, people loved playing multitable tournaments online. When we had a soft opening for real money in December, the movement of players to start playing for money was just that: soft. We were the largest play money cardroom, but struggling to get real money critical mass. The root of the problem was simple... our play money games were too good!

Usually play money games are a joke because one or more players just raise or bet every time. But this is a lot less prevalent during multitable tournaments, where the unserious players go broke quickly and can't get back in.

So, January-February 2002 we had endless discussions on how to convert our ocean of play money players into a decent pool of real money ones. The timing of these discussions occurred simultaneous with the negotiations with Chris Ferguson to become the face of PokerStars. In the linked article I mention that a deal wasn't made with Chris partly because of disagreement over how compensation would be paid, but also because of a difference in business philosophy. Chris believed in a much more laissez-faire, "hands off" philosophy. (This hands-off versus hands-on difference is the root of the difference in the business choices PokerStars made compared to Full Tilt years later.)

My hands-on views are evident in what I proposed as one element of transitioning play money players to real money ones. I advocated that all play money big bet games have 1/2 blinds, and all limit games have 10/20 blinds. Forever.

Since it is common now for some play money players to have over one billion in play money on a table at one time (and billions more in their accounts), I sometimes imagine the savage hordes of play money players hunting the earth to lynch me for forcing them to still play with 1/2 blinds...

In retrospect, prior to Black Friday I believe I was right. Permanent play money players may not be that harmful (they do waste a significant amount of support personnel time) but after a decade of not converting to real money, they don't add any meaningful value either. Post-Black Friday though there is some value in having tiers of play money games to give United States players a reason to keep playing on the site.

Why is it the WCOOP and not the WSOOP?
Once PokerStars opened, it was only a matter of time before the other online cardrooms would figure out how to offer multitable tournaments. Knowing this, from the beginning it was my highest priority for PokerStars to launch the first worldwide multitable tournament series, and make a big deal about it.

After slowly overcoming the hurdle of converting a critical mass of our players from play money to real money, we decided this first tournament series should occur in June, since the 2002 WSOP would conclude in May in Las Vegas. So we knew the timing, settled on the events, but that left: what do we call this thing?

The World Series of Poker was still owned at the time by the financially strapped Behnen family (Harrahs would not buy them out until 2004). Since the Horseshoe never got consent from the baseball World Series to establish the WSOP, why not just establish our World Series of Online Poker? On the other hand, why solicit a confrontation with some legal firm who would convince the failing family company that getting litigious was a no risk option for them?

We decided to hedge our bets. We chose to use the World Championship of Online Poker brand, but trademark the names World Series of Online Poker and World Series of Internet Poker, though not use either initially. Without any legal drama the first WCOOP was a great success, and continues to be one. In contrast, when UltimateBet predictably announced they were offering a tournament series called the World Series of Online Poker, PokerStars' attorneys contacted them with our trademark information, and I assume the Behnen's contacted them too, because UB ditched the "World Series" name almost immediately.

PokerStars subsequently abandoned the trademarks due to lack of use, so presumably if ever Harrahs expands their online poker offerings worldwide they will make use of the World Series of Online Poker title, but to date they only offer WSOP online bracelet events and online poker play in a handful of US states.

Finding a cardroom manager
In the beginning, one way to attempt to be different from Paradise Poker was to hire known poker people. Two of the people I early on recommended for marketing jobs, Dan Goldman and Rich Korbin, were known players (Rich was even a WSOP bracelet winner). But for the job of cardroom manager we were looking for someone more from the industry side of the business. I asked a couple people from the Commerce Casino; both said thanks but no thanks. Same with two people from the Bicycle Casino, one from Hollywood Park, one from Reno and one from Bay 101. Nobody with a good casino job was looking to hitch their wagon to a start-up online room.

So, since he was currently working as an independent tournament director, I suggested Steve Morrow for the job. Eventually he agreed, but after awhile it became clear it wasn't a good match so Steve left and we had to find someone new. Up to this point, everyone seriously considered was someone I knew. We even went back to the Commerce/Bike/Bay folks to be turned down again. Then Isai asked me about Lee Jones' reputation. I said I didn't know Lee, but his book was mostly well-regarded (except by Mason...), and he was from northern California and I would always support hiring anyone from from NorCal. While southern Californians might fold a hand pre-flop occasionally if they have to use the restroom or something, northern Californians feel a responsibility to play every hand they are dealt in -- and sometimes they even play when they have not been dealt in!

The worst of the worst ideas
Every business will do dumb things once in awhile. Some businesses, like the old Full Tilt Poker, will do endless amounts of dumb things and still manage to stay in business for years. Sometimes the dumb business decisions of competitors will cause otherwise sensible people to follow the same dumb path. I mention in the sponsored players article how in 2009 my head almost exploded at the ludicrous seven-figure numbers being thrown around as a high profile player was negotiating with both PokerStars and the old Full Tilt to become sponsored. My view was he wasn't worth 10% of what we were offering him in response to what Full Tilt was offering him.

But even that brain hemorrhage was dwarfed by this whopper: In 2008, serious consideration was given to paying $10,000,000 to have the PokerStars logo on all the World Series of Poker tables. In response I said putting that ten million into 2+2 or other online advertising would generate essentially infinitely more new players. While the logo on a table might make sense in 2004, by 2008 every player was wearing a patch for one cardroom or another -- and we were already the largest cardroom with the largest brand awareness. I don't think a table logo would have even generated 100 new players, let alone the 10,000 or so needed to make it a profitable idea.

Happily, we didn't go down this rabbit hole, but the much smaller Everest Poker did, and it basically destroyed them.

While that $10 million tabletop idea was the worst I ever heard, an even bigger number was thrown around for a more intriguing asset: in 2007, the domain was put up for auction, with a minimum bid of $22,000,000. Domain names have little value in themselves, but if PokerStars, as the industry leader, owned the one word poker domain, it would be a branding no-brainer. For $5 million, we would have snatched it up, but for over $20 million...? As of 2018, it remains for sale at $20,000,000.

It's the Poker, Stupid
As I write in The Poker Ecosystem, one reason PokerStars thrived while many competitors failed is because we (almost) always kept as a core principle that the #1 draw of the most successful poker site must be the poker. Our competitors flailed around offering one gimmicky marketing strategy, bonus code or rakeback formula after another. In contrast, the original PokerStars team aimed to make the best online poker software for poker players to play on. That's it. That was the "genius". Make better software. Make better functioning games.

Sites run by non-poker marketing people never dominate the market because it's the poker that matters, not the wrapping paper.

On the other hand, marketing should follow the lead of a sensible business model. With PokerStars, the first thing was "building a better mousetrap" than Paradise Poker. (It's the poker software, stupid!) Only then should you prioritize the next thing, getting your superior software and game experience sampled by large numbers of people, especially new people, who will judge for themselves.

The old Full Tilt Poker failed because of poor corporate vision and very bad practices. One of their worst "vision" failures was how they targeted the USA almost exclusively. One of the things I'm most proud of in my association with PokerStars was in getting the company to translate its content and market to every major language. Time demonstrated that a company with a worldwide player base was superior to one that only focused on a single country, but that "proud" comes with an asterisk... since it took me three and half years to convince the company that translating content was vital!

It seems absurd now, but it wasn't until 2006, when PokerStars hired Randy Ray to work with me (Randy did the work, I did the talking) to improve the search engine rankings for the PokerStars family of websites, that translations were done -- and even then, Randy initially paid for them from his own pocket to speed things up! In two months, search traffic to the Stars websites tripled. More about that when I write my Poker SEO article...

When Paradise Poker was the industry leader, they did things better than the competition. When Party Poker was the market leader, their software wasn't quite as good as the best but their marketing (via exclusive World Poker Tour commercials) was greatly superior to the competition. With PokerStars, the mission was to do things right and do the poker things better than anyone else. If another cardroom ever wants to surpass PokerStars, they likewise need to not just do things better and do things right, but they will also need to demonstrate that they understand that it is the poker that matters.

See also Amarillo Slim and PokerStars, How PokerStars Got Its Name and Stories from the Poker Boom