"Ironically, your profit would shrink to zero if the site stopped collecting rake! The thing that more than any other makes games tough to beat - the rake - makes it possible for your bot to turn a healthy profit. And that, as a long-time online poker player, makes me smile."
That's not true at all. In fact, the logic is terrible. If your bot is breaking even at a raked table, and you remove the rake, it would become a winning player. Far more so, in fact, than it makes just taking 35% of the rake back.
I definitely agree with you, but I was trying to think about how what he said could be true. Is it possible rakeless games would alter the strategies of the opposing players enough to render the bot ineffective?
But I guess even then the bot could be modified accordingly. Interested to hear the author's response...
It would alter strategy a little. You should play more hands. There are a number of hands that are very close to 0 in EV that become positive EV with no rake. Not that many though, especially in low limits.
I'm sure the bot would be modified accordingly. My guess is it would add a couple percent more hands in. I'd further suspect that the increase in postflop play would then lower the bot's expectation (bot's are incredibly well suited for pre-flop play, not so much after) so though it would still do better than at a raked table it wouldn't be the 3x you'd expect.
It would do better though. The skill differential would be nowhere near large enough to overcome adding a hundred or more dollars per hour to the table. It'd still probably be at least 2x I'd guess.
The bot would also make more in the long run because the people it makes its money from would go broke slower.
It's possible, but doubtful. I would assume that people who pay a lot of attention to the rake might become a little more loose, since the rake is essentially a tax on hand activity.
Looser players would help a poker bot, instead of hurting it, because a poker bot never gets bored, and is content playing winning strategies 100% of the time (i.e. it doesn't feel like it's going to get lucky playing KJ off "this time").
It seems to me that the "rakeback" is a promotion that will give you some of the rack back. This case if you are breaking even with the rake you are going to get some of it back. Otherwise the author doesn't make sense at all for the reasons you mentioned.
That is what he is saying, he just doesn't understand that if he took the rake out, the bot would be doing much better than breaking even. And much better than breaking even plus 30% rake back at a raked game. About 3x better (not counting any strategic adjustments).
I thought that the topic was covered comprehensively last week ( http://news.ycombinator.com/item?id=185830 ) but, there's plenty to discuss about privacy invasion and bot counter-measures.
That's not true at all. In fact, the logic is terrible. If your bot is breaking even at a raked table, and you remove the rake, it would become a winning player. Far more so, in fact, than it makes just taking 35% of the rake back.