From Chapter 90 of 100 Essential Tips to Master No-Limit Hold'em:

One problem with satellite tournaments tournaments is that your ROI is capped. In a normal tournament, if you win, it will usually be for 30 times your buy-in or more, but in satellites, the most you can win is 10 buy-ins (or perhaps more, depending on the structure). This will lead to a lower average ROI because a large amount of a strong player's edge comes from getting to the final table with a large stack and then winning it for a lot of buy-ins.

All comments welcome.

Mason

I'm not sure if there was another paragraph that followed and elaborated in more detail about what he was talking about?

Being familiar with JL's work, I have heard him criticize satellites for being too simple. The best strategy is often to build up a big enough stack, and then just play extremely tight until the bubble bursts.

With traditional tournaments the changing ICM dynamics, especially on the final table, create a lot more complex situations that can lead to a bigger edge for good players.

I don't want to put words in his mouth, but I suspect that is what he meant. I tend to agree with that particular part of his perspective.

However I don't necessarily agree that a "cap" on how much you can win in satellites is a bad thing. The structure also likely lowers variance, which is a good thing.

I put this up because from reading this paragraph it's clear to me that Little has a very poor understanding of basic mathematical expectation. Real quick, it you can win 10 big blinds more than three times as often 30 big blinds your expectation is probably higher in the satellite. It's actually more complex than this because in a tournament there are other payoffs as well.

Now, to be fair, a lot of good poker players don't understand this stuff very well and over the years I've seen numerous errors like this. But I do think that if you're writing and selling information about poker, as well as selling courses and making instructional videos, your understanding of mathematical expectation should be much better than this paragraph indicates,

Thanks for the response,
Mason

The ROI concept is not a good framework for thinking about poker EV unless multitabling online. Or if you're a staker with a large stable, rather than one of the horses. "Return on investment" presumes that the investor can manage many investments simultaneously. Hourly rate is more appropriate if you can only sit at one table at a time. Even if you have 100% of your own action, you're basically an entrepreneur rather than a capitalist. Your own "labor" is still essential.

A gambler's time is limited, and not worthless. If you're in a casino, presumably you can find some other +EV use of your time after you cash or bust out of a tournament. A 4 hour satellite with a $100 EV is worth at least $25 an hour. If you're playing a longer tourney where you have to wait until hour 8 to cash at all, your EV would have to be at least $200 to match that hourly. Maybe a lot higher.

Has anyone come up with a version of the Kelly criterion for the player's "time roll?" That would be relevant here.