Lessons learned from a failed poker software business

http://www.mattmazur.com/2008/11/all-in-expert-lessons-learned/

by matt1 • 18 years ago

103 66 18 years ago

nadim• 18 years ago

Thanks for sharing the insight. A few months of work in exchange for some valuable lessons learned - well done.

Does "estimating your opponent's hands" have any value? If you could do that with any certainty you probably don't need a calculator.

EDIT: I realize that estimating your opponent's hands is important. Basic poker there. I'm talking about "estimating your opponent's hands" which is featured in one of the calculator's screens. That seems like an iffy thing to feed into a computer. It's a garbage-in-garbage-out situation.

jd3• 18 years ago

Being able to figure out what kind of hand your opponent has is the difference between winning and losing poker. You usually make a guess based on previous behavior, how much aggression your opponent has shown so far, the position your opponent is in (acting first is a huge disadvantage) and so on.

If you make no estimate what kind of hand your opponent has you can only reason in three ways:

a) you assume he always has the best hand. So you throw away your hand.

b) you assume he always has the worst hand. So you put your chips in.

c) you bet big if you have good cards, and you throw your cards away when your cards are lousy.

Clearly (a) and (b) are bad strategies.

The last strategy is not as bad as the other two, but still terrible. More often than not your cards are bad. So your opponent will figure out you're not willing to make large bets with lousy cards, so every time you don't bet, he will, and every time you bet, he won't. You're going to lose your chips that way. Basic game theory.

So you need to figure out what cards he could have been dealt so that his actions make sense. Then, from that range of plausible cards, figure out how often he has a better hand than you do. If you like your odds, you put your chips in. Otherwise, you throw your cards away.

It's a lot more complex than this, of course. But the point is that "estimating your opponent's hands", as difficult as it may be, is absolutely crucial if you want to win.

matt113• 18 years ago

Great post

Retric• 18 years ago

I suspect that the best long term stable strategie is to randomly assign a personal value for this hand. Something like .75 to 2x the actual value of your hand before the bet to reduce the amount of signal your sending out. If your both doing this and playing perfect poker then nobody wins.

Finding what the best random function might be would be an interesting test of machine learning. But in the real world few players are going to do something like this so finding out what their approach is is going to be a better approach.

matt113• 18 years ago

Different players think of different levels.

A beginner will primarily focus on what his own cards are. An intermediate will focus on what his opponent's cards are. A more advanced player will focus on what his opponent thinks his cards are and so on. The best players in the world are experts at figuring out which level you're thinking on so they can think one level beyond you.

Once you estimate your opponent's range based on his style, the cards, his previous actions, etc, you need to know what to do with that information. ALL IN Expert was designed to help with the second part.

mattmaroon11• 18 years ago

Estimating your opponent's hands I assume means guessing what range of holdings he would play that way. For instance, if I tight player raises under the gun, you can narrow down his hands to just a few.

It has immense value, in fact, it's the hardest part of the game. Figuring out what to do once you've estimated someone's hand is (with a little practice and the help of a good computer program) pretty easy.

The entire game is one big war to see who can estimate his opponents' hands the best, while preventing his opponents from doing the same to him.

vlad• 18 years ago

I don't think you had a failed poker software business. You wrote, released, and announced some software. That's simply step one, and the very minimum. Next, you need to iterate, which may include adding features found in other software, creating a web version, and more. I see very few lessons here, and they are on how to fail.

matt113• 18 years ago

Yep, you're right. As I mentioned in the post, I use the word "business" liberally for lack of a better term. If there had been a future for the product I would have iterated, created a web app, etc, but I decided it wasn't worth the effort. As far as the lessons, figuring out how something fails is strongly tied to figuring out why it succeeds.

breck• 18 years ago

"but I decided it wasn't worth the effort."

Good decision. It seems like you have a lot of talent and can make a LOT more money in a vertical other than poker. Also, it will probably be more enjoyable in another domain.

My 2 cents. We tried to a business in the lottery space once because it seemed there were no good tech companies there, there was a ton of money at stake, and lots of innovation to be done. But after a short time it was really demoralizing to be in that type of vertical. Fail fast and move on!

patio11• 18 years ago

I have a small software business, and I always say that 95% of the business value is not produced in the compiler. Its marketing, support, marketing, SEO, marketing, advertising, marketing, documentation, marketing, etc.

Oh, and there is some software involved, too. But product quality only goes so far if your strategy for marketing is "I put it up on the Internet and eventually people find me". (That _can_ work but you have to have God's own luck AND an amazing product.)

Here's another lesson for you: if people are not buying your software, the answer is NOT to charge less. You can't convince people who aren't at your site to buy by offering them a discount -- they can't see the discount. On the other hand, folks who then see the discounted price as the "normal" price mentally value it to match. And software which only costs $10 is worthless. You are selling me a tool to _make money_ and can't justify a price higher than $10?

(I leave it to your imagination what I think of the "Make it free and try monetizing with ads!" advice you'll find elsewhere in this discussion.)

I sell software which is so trivial that technical types have difficulty believing it sells at all. My customers love me, and are happy to part with $25. And I should really get around to raising that one of these days. Should sell about $20k this year, which isn't quite quit-the-day-job money but on an hourly basis works out to "pretty freaking sweet". Not to crimp the startup style here but, erm, charging people money for value remains a decent business model.

Tichy7• 18 years ago

It sounds as if you didn't try any advertising at all, except for a small announcement in a poker group? I think you should have put a lot more thought into how to get users to find your product. Like, sometimes (often, maybe), people don't even know they could use something and that something they could use exists. You have to educate them.

matt113• 18 years ago

Good point. I added a bit of clarification to the article since it did read like the one post was the only thing I did.

After it launched I decided that the best way to demonstrate its utility was to search poker forums for problems and then post the ALL IN Expert results. I wound up adding a feature to the software that let people export the results in a forum-friendly format which also linked back to the site. This helped bring some people to the site and if I had continued it for some time it may have brought in a few more registrations. In hindsight maybe I should have done this more. It's hard to say... No matter how much you prod a dead duck it's not gonna move much.

jd3• 18 years ago

Were you aware of the existing (and immensely popular) free tools when you started programming?

If so, what was going to be your edge?

Tichy7• 18 years ago

Not saying that it would have been successful, hard to say.

But reading "Advertising is a tax for being unremarkable” seems a bit harsh. In some businesses, there simply is no other way. For example recently I heard from some online dating services that they really boil down to the hard calculation of "how expensive is it to convert a user we bought with advertising". Like if they raise their conversion rates, they can afford more advertising, and so on. Sure, maybe sometimes there will be a "hit dating service" that spreads through word of mouth, but the others seem to do OK, too.

Especially in Poker it seems advertising is huge?

halo• 18 years ago

I note that on the website there seems to be no real screenshots of your application in action that give me a good idea of what it does and how it works. In my opinion, this is a huge mistake - I generally won't download /anything/ unless I've seen a screenshot first.

matt113• 18 years ago

You mean the little dinky gradient grid on the front page isn't enough? Just kidding -- thanks for the feedback.

lazyant2• 18 years ago

Perhaps you can make an online free version and put some ads?

matt113• 18 years ago

Given the small target audience, I don't think it would be worth it to convert it into a free web app. Anyone that wants to toy with it is welcomed to download the desktop app now anyway.

rantfoil• 18 years ago

I don't think that's true at all. It depends on how much use you really can drive. With an online app, with no app install, and purely organic growth and a whole ton of community evangelism, I could see it be a really successful web app.

Aren't online gambling sites some of the biggest spenders on online ad sales? With revenue of $6+ billion per year, I think a good chunk of that must go to lead generation.

lazyant2• 18 years ago

I was mentioning it because there may be people interested in a web version of the poker calculators. As far as I know the known names (pokerstove etc) are all Windows-based and only Ed Miller has an online equity calculator in his page.

There may be more poker players like me who doesn't use Windows at home and I could see the value of having a re-sized small web window while playing with just three boxes (pot, my stack, opponent's stack) plus a small pre-selection of opponents ranges (ex: top 10%, top 40%, any two) so it can be used in real time.

mattmaroon11• 18 years ago

Poker players are pretty used to installing GUIs, so I don't think web-based would help. Every poker client is a GUI (and for good reason), as is Poker Tracker, Poker Stove, SNGWizard, etc.

I don't know if you're doing ICM calcs or what, but anything beyond a simple odds calculator should be a native app.

Fuca• 18 years ago

How can this be a Fail? Give it for free if that what it takes then adapt it to make money.

matt113• 18 years ago

"Give it for free if that what it takes then adapt it to make money." -- I laughed.

With a lot of work it might one day make some decent change, but again, it's not worth it at this point.

time_management2• 18 years ago

I wouldn't consider this a failure. Guessing that you don't want to run into legally shady territory (assuming you're in the US) and start an online gambling site, it's hard as hell to make money in anything related to games, especially public-domain card games. For example, I'd doubt that the designers of card games like FreeCell and Ambition ever made a cent off their inventions.

I'll point this out, though: games do well when the economy tanks, so I wouldn't give it up just yet. (Monopoly was invented, in its modern form, during the Great Depression.)

mattmaroon11• 18 years ago

You have no idea how wrong you are. There's a tremendous market for poker-related apps. It's an industry that rakes in at least hundreds of millions annually.

If I told you what I made off of this sort of thing in the pre-UIGEA days (and especially before Party spun off its skins sites) you'd shit your pants.

Top rakeback providers made 6 or maybe even low 7 figures per month at the pinnacle. The Poker Tracker guys have sold many thousands of copies at $50 a pop.

Games in general are the only thing monetizing well right now on Facebook and MySpace platforms too, and the #1 most profitable is almost certainly Zynga's Texas Hold'em.

time_management2• 18 years ago

I looked into it a little bit. You're completely right. I was wrong.

I remember looking at the casual games business in 2006 and realizing that there was almost no money in it at the time, but that changed, and Texas Hold 'em is, anyway, it's own beast.

I'm very surprised, though, that the Texas Hold 'em blip is still sustaining. It's a fine game, but you'd think people would get sick of playing the same game after a while, and look into something more adventurous.

alexhu• 18 years ago

I'm guessing your problem is a miscalculation in target market.

People will be either

1: Very good at poker. In which case they can go through this mental, odds-calculating process in their head without the need for software. or

2: Not very good at poker. In which case your software is immensely confusing. They need to be able to put opponents on a range of hands (how does a novice guess what an opponent most likely has? half-likely has?) in less than 20 seconds (the time online poker software gives to each player per turn).

I play poker pretty often, and can tell you that odds calculation is never a fixed thing (so it's hard to replicate with software), and even considering my poker background, the screenshots of your program still confused me.

mattmaroon11• 18 years ago

1) Wrong. Complex odds can't be calculated in your head. Even the math PhDs who do well at poker calculate EVs against a range of holdings with a computer.

For instance, if you can calculate your equity with T9s against a range of (AA-22, Ak-AT, A9s, KQ-KT, K9s, QJ-QTs)in your head, consider yourself a genius.

The type of software he built is very popular among people who are good at poker or trying to become so.

brent2• 18 years ago

$10 (or $40) isn't enough for me to take you serious. Would you make a move on $10 stock advice?

mattmaroon11• 18 years ago

I would buy a $10 program that aided me in making stock picks if it were simply doing mathematical computations for me. This is more the equivalent of a program that computes PE and dividend ratios than one that says "Buy AAPL at $95". Except that dividend ratios can be computed trivially in your head, while equity against a range of holdings given certain chip stacks cannot.

brent2• 18 years ago

Point taken. I didn't read the details. Sounds more like it truly is worth $0.

rms3• 18 years ago

Why don't you give it away for free to people who sign up using your affiliate link for a poker room?

mattmaroon11• 18 years ago

Hard to make any money that way. Most people who want this type of program already have accounts at the one or two sites worth playing at.

rms3• 18 years ago

I was thinking of this program, which is distributed that way. http://www.calculatem.com/ There are a lot of casual users who would definitely be interested in something to give them an edge, but I guess the problem is they don't know such programs exist.

And sure, most people probably already have accounts at their preferred sites, but it doesn't hurt to have it as an option. It's easy enough to open a new poker account (well, it was when I was playing before Bush messed things up) and getting the typical bonus + software could be enough to play for a week at a second tier site.

mikkom• 18 years ago

I think taking down the site was a bad idea, after reading your article, I would have bought it (my pain treshold for random purchases goes somewhere around $20 so if you decide to sell it for under that, please drop a notice).

I'm an entrepreneur myself (We sold our company to a quite big company about 1,5 years ago), I think what really went wrong was not the product, niche or anything like that. It was the marketing/sales. You should never underestimate how hard selling even a good product is.

also, this kind of a product will never "get old", the principles behind it will be the same forever so if it's good and you get enough people who purchase, you will get snowball effect by mouth to mouth marketing.

I think the worst decision you made was to waste your work when you took the site down. I also think your name is awful, it's so generic that it's hard to even find using google so what I would do is to republish the program with catchy name and add a link from your blog.

Also it's not a bad idea to hang in places like digitalpoint forums and hire some indian to submit your tool to X directories, that will cost you about $10-20 and you will get a lot of google goodness.

edit: you are giving it free now? Okay suits me but I would have bought it. How about adding at least a donation button?

matt113• 18 years ago

Very good post. I have a lot to learn about advertising. While it's nice when your product sells itself, there's clearly a lot of value in a great advertising campaign. I could have done much better following through after I launched.

I liked the name, but the problem was the three words: all+in+expert. There was a time after it launched where the site came up as #1, but now its faded into oblivion. I thought it succinctly captured what the software was about, but as you said, maybe it wasn't catchy enough.

I'm launching a new site in a few weeks which I'm going to do a few things differently with. The digitalpoint idea isn't bad.

No need for a donation, your advice is far more valuable than $20 will ever be.

dhimes• 18 years ago

Thanks for posting. I'm finding myself in a similar situation, but I've adopted a different strategy. I'm writing a second, semi-related app that I hope will help build the brand equity. There are a lot of insightful comments in this thread.

My marketing and advertising are weak.

daemon• 18 years ago

You would have generated the target 40K in profit if you'd have simply taken the time you spent on developing the odds application, and put it into building a poker bot and playing limit hold'em instead.

Lessons learned from a failed poker software business

Related Stories