Does this mean that games like Tropix and Escape From Paradise, which combine all the typical Reflexive Arcade games (Match-3, Hidden Object, Diner Dash, etc) into one sell the product of the amount of clones they've included, or do they sell as much as a game that clones just one concept?

I think you mean 19 times as many copies, not 19 more copies.

Anyway, this argument is pretty subtle, since it's not about whether clones are good or bad, it's whether they sell better. I'm not convinced that that's the only interpretation of the data he mentions.

To really get an idea whether clones sell better than original games, you have to compare best-sellers which were clones to best-sellers which were original, and see which sold more, you can't simply look at all games that are selling and say that because most games that are sold are clones that that means clones sell better.

In other words, I suspect original, genre-forging or genre-innovating games sell better than the games that clone that original game. Pac Man sold better than all the maze game clones. Tetris sold better than all the dropping block clones. Dragon Warrior sold better than all the RPG clones (except for Final Fantasy, which innovated in story presentation in any case). There are more examples, but the point is that if you're cloning a game, a lot of the time the game that is cloned-but-original sold better than all of its clones, sometimes better than all of its clones combined.

It's a different point Paul, but one we're working on for the next article, how the original sells in comparison to the clone, but the question quickly comes up of what is the original. Is Tapper, Betty's Beer Bar or Diner Dash the original? How about the I Spy games or Mystery Case Files? Still we're running the numbers to see if there is anything interesting to talk about. Certainly Betty's Beer Bar didn't sell better than all the clones combined, in fact it didn't sell as well as most of them.

To the other point, comparing best sellers to best-sellers, I'm not sure what it would tell, which best selling game sold best?

For example, Virtual Villagers sold VERY well. However, that doesn't say much about the average 'different' game any more than Mystery Case Files says about the average 'I Spy' game. The thing that is interesting to me is that even when you include VV along with all the average games that are 'different' they sell at a much worse ratio than things that aren't different.

And you are right, it's not about good or bad. That's a different discussion. And the numbers can tell different stories...I'm hoping for an interesting one next month as we keep doing different queries against the numbers :).

This sort of thing is tricky, you could interpret it in almost any way -- for instance, maybe Reflexive are just way better at marketing clones than original concepts.

The fact that this addresses casual gamers specifically makes the results very unsurprising; the equivalent would be Macdonalds comparing sales of their regular menu to an experimental "MacHauteCuisine" menu.. of COURSE casual gamers prefer clones, all they care about is stuffing their brains with instant-gratification junk food. It's like discovering that 9 out of 10 alcoholics prefer hard liquor to fine wine.. they're not drinking it for any reason other than to get hammered, so it's not surprising that they don't care about the bouquet/etc.

This is why the only "solution" to the casual market is to boycott until they become literate or involved enough to enjoy proper games.

Raigan has a good point.

I think that determining the original game can be tricky, but it's not impossible. For instance, Dragon Warrior is clearly the first Japanese-style RPG and the one that was cloned by all the others. There were RPGs before it, like Ultima, but they played very differently, Dragon Warrior simplified and organized it in the form that most games cloned. So I think that just because it's often difficult to determine whether a game is a clone or merely similar, that doesn't mean we should throw up our hands and say we can't tell the difference.

And what I mean about best-sellers is this: because most games sell so poorly, particularly indie games, you can only really judge trends on the best-sellers, the vast majority of games sell in too small numbers to be statistically significant; this is known as the rule of small numbers.

For instance, if my game Immortal Defense so far sold 150 copies, that doesn't say anything about abstract games in general or tower defense games in general, the numbers are too small to judge, it isn't a good basis of understanding when the data is that sparse. So only by comparing the best-selling games, and not the vast majority of games, can the data be relied upon.

And even if this were done, the information obtained (whether clones sold better or not) wouldn't be that important to the developer, because the key thing that matters from the business standpoint is making a best-selling game, not whether that game is going to sell well for a best-selling game.

In other words, the data you used doesn't seem to me to mean that clones are any more likely than not to become best-sellers, it only means that more clones sell better on average. The reason they do may simply be because more of them are made. Or it could be because people are more likely to clone a good game than a bad game and good gameplay sells better than bad gameplay; whereas original games are more likely to have bad gameplay (not because they are original, but because they didn't clone good gameplay). Or it could mean that people who make clones have more time and energy to invest in the graphics and sound and polish, and it's graphics and sound and polish that sell games more than gameplay. There are a lot of alternative explanations other than the one the article gives.