next »

[airman4]

i don't agree with "the best games always sell well" -- i can think of some very good games which sold very poorly. examples:

psychonauts
planescape torment
ico

i believe those all sold less than 100,000 copies

I total agree

It's like saying thatv Call of or Just dance are just greater games than the most since they sell well.

I mean okay , call of is a great game but come on , if all the best selling games were truly good Damn ...

(and yes , some poor games sell poor)

[DIT]

I think good games need good marketing (or press coverage, or community support) to sell.
Also good marketing requires a game that is at least a decent product.

You also need to ba able to get you game on distribution platforms that are actually able to sell something.

For istance, if you build a pc game and you are not on steam, selling the game is going to be 100x harder.

There are several mixed things that influece your sales, the *pure* quality of the game is just one of them... sadly.

[BlueSweatshirt]

metacritic

Seriously, Paul?
How about because there's a huge grey area in metacritic's rating system.
Oh, and, also that lots of "professional" game reviews have artificially inflated scores.

CONSIDER THE INTEGRITY OF YOUR DATA SOURCES.

[Blademasterbobo]

how many people in this forum cannot actualyl read this is a s eirous dileman cugys !!  it'js  not  hard do you jus t gotta     s e e the words  1  ! and then . . . .  t he r es t is up to you ! 1

[Paul Eres]

metacritic

Seriously, Paul?
How about because there's a huge grey area in metacritic's rating system.
Oh, and, also that lots of "professional" game reviews have artificially inflated scores.

CONSIDER THE INTEGRITY OF YOUR DATA SOURCES.

it's not like i personally chose which standard the study used for its review scores, not sure why you're getting so upset

but i do think that metacritic is more accurate than any other one site's review scores, because it's an average of all the other review scores. i wasn't saying that metacritic reflects how good a game actually is, but it does reflect how well a game did in reviews

the study showed there was no correlation between how well a game does in reviews and sales, after a score of 70. that's accurate whether or not you feel that games are reviewed fairly

[Pandara_RA!]

metacritic

Seriously, Paul?
How about because there's a huge grey area in metacritic's rating system.
Oh, and, also that lots of "professional" game reviews have artificially inflated scores.

CONSIDER THE INTEGRITY OF YOUR DATA SOURCES.

it's not like i personally chose which standard the study used for its review scores, not sure why you're getting so upset

but i do think that metacritic is more accurate than any other one site's review scores, because it's an average of all the other review scores. i wasn't saying that metacritic reflects how good a game actually is, but it does reflect how well a game did in reviews

the study showed there was no correlation between how well a game does in reviews and sales, after a score of 70. that's accurate whether or not you feel that games are reviewed fairly

Link to study? I'm interested if they polled a large sum of games across all time periods (LoD and such getting on there or say harvest moon would skew results as gaming is a growng and expanding trend so even some of the "best games" may not have sold 1,000,000 copies. Or maybe the success is based off the cost of making the game?

Is it taking indie games along for the ride in that study too?
I'm sure some of the greatest indie games didn't sell too well, but got super high scores, again, screwing with the results by being outliers.

[Paul Eres]

it's been a while since i read it so i don't remember it exactly, but on googling, i think this was it: http://bitmob.com/articles/the-influence-of-metacritic-on-game-sales

[Paul Eres]

note that the second charge is a "zoomed in" view of the first chart (different scale on the bottom)

you can see from those charts a few things:

1) *most* games sell badly, even games that get metacritic scores in the 90s sell badly sometimes (the vast majority of games in that chart sold less than 200,000 copies)

2) of the games that sell very well, they tend to be a bit random (strikes of lightning)

3) in the zoomed in portion, you can see a *rough* correlation, where games below the 70s sell generally poorer than games above the 70s (ignoring outliers)

so basically, a game sucking does hurt its sales, but not as much as you'd think, and only if it *really* sucks. but there are still plenty of games with metacritic scores in the 30s that outsell games with metacritic scores in the 80s

[Pandara_RA!]

note that the second charge is a "zoomed in" view of the first chart (different scale on the bottom)

you can see from those charts a few things:

1) *most* games sell badly, even games that get metacritic scores in the 90s sell badly sometimes (the vast majority of games in that chart sold less than 200,000 copies)

2) of the games that sell very well, they tend to be a bit random (strikes of lightning)

3) in the zoomed in portion, you can see a *rough* correlation, where games below the 70s sell generally poorer than games above the 70s (ignoring outliers)

so basically, a game sucking does hurt its sales, but not as much as you'd think, and only if it *really* sucks. but there are still plenty of games with metacritic scores in the 30s that outsell games with metacritic scores in the 80s

you'll notice the games that sell are the ones with the big publishers behind them, there aren't enough names given for a good reference but from what I'm seeing if anything that it takes a "decent game" + Big marketing = Success. Also the demographic of the games are a HUGE difference. An adult adventure game with crude humor is so niche, same with boom blocks a party puzzle game.

compare that to sports (sports+fitness) X-men (um it's X-men, tons of kids and adults know of that name) and so on. I think there are too many factors that could go into this chart to come away with "rating doesn't = sales" because the graph is comparing apples to oranges to peaches to plums. If it was Action game with strong marketing vs action game with strong marketing + good score we might have a case...but still an iffy one. I bet boom blocks having a high score would result in way more sales than say another niche party puzzle game for example.

Comes down to marketing in the end like selling sand to someone in a desert.

[Paul Eres]

it's possible the sample size isn't that great, but certainly there are a lot of other factors that differ, but the way statistics work is that you are always going to get lots of factors that differ

example: let's say there's a study of exercise vs longevity (whether exercising makes you live longer). the individual people in that study are going to differ in many factors which affect longevity: genes, wealth, health care, diet, and tons of other stuff. but we don't usually say that the results are invalid just because it didn't take into account all of those factors

[Pandara_RA!]

it's possible the sample size isn't that great, but certainly there are a lot of other factors that differ, but the way statistics work is that you are always going to get lots of factors that differ

example: let's say there's a study of exercise vs longevity (whether exercising makes you live longer). the individual people in that study are going to differ in many factors which affect longevity: genes, wealth, health care, diet, and tons of other stuff. but we don't usually say that the results are invalid just because it didn't take into account all of those factors

We don't usually say the results are invalid because the sample size in that study is actually so insanely large even with outliers the results come in clear.

For these graphs here we are sampling a "class room size" of about 20-30ish games. Not a good sample size so the results are not very trust worthy and  it's why the factors of things like large marketing and niche roles (outliers) play a much larger role than they should in telling the story about game rating not affecting sales.

[Paul Eres]

yeah, i think it'd have been better if he did this for every month of a year, to see if the correlation holds up in a larger sample size

[James Coote]

A lot of it is that developers don't enjoy marketing, even if they recognise it is necessary to the game's success

That means they don't go the extra mile when it comes to marketing, and as pointed out, the marketing is half the fight

[Fallsburg]

Well, coming from a statistics standpoint, I can think of a lot of factors that you would want to control for if you were doing a serious analysis.

Off the top of my head:
-Game Budget
-Marketing Budget
-Publisher Recognition
-Developer Recognition
-Franchise recognition

And all sorts of cross effects. 

Now, this would be a serious undertaking (as you would have to run separate studies for the recognition effects) and might be impossible (not really sure how much is disclosed about game and marketing budgets).

[Pandara_RA!]

Well, coming from a statistics standpoint, I can think of a lot of factors that you would want to control for if you were doing a serious analysis.

Off the top of my head:
-Game Budget
-Marketing Budget
-Publisher Recognition
-Developer Recognition
-Franchise recognition

And all sorts of cross effects. 

Now, this would be a serious undertaking (as you would have to run separate studies for the recognition effects) and might be impossible (not really sure how much is disclosed about game and marketing budgets).

Exactly what I was thinking, which is why I found it really silly to have a study with a classroom size of 20 game and even then it's a very non-scientific opinion of the outcome of those results.

Real shame the NPD sales reports are sold and not public =[. I'd love for the study to be solid.