Standard deviation of game reviews

[Jagulars]

There are games that get mixed reviews because they are "meh" and people generally don't like or dislike them.

There are also games that get mixed reviews because some players love them and some players hate them.

How can you know the difference?

EDIT: It seems I need to explain myself better. There are thousands of games. Few games are worth checking for me. Personally I trust user reviews more than a pretty facade. Consequently, I check only ones that have "overwhelmingly" or "very positive" reviews. This causes me to miss games that I could like because they are targeted to my niche but have "mixed" or "mostly positive" reviews. I don't have time or interest to delve deeper into every game to see if it's worth my time. Who does?

When we go shopping we check the surface and maybe inspect the item more closely if it attracts us. Nobody checks every item in the supermarket thoroughly before deciding what to buy, for obvious reasons. We need filters for what we pay more attention to. Natural conservation of energy.

The more accurate information we can get from a product beforehand, the less is left for hunches and guesswork and the less we waste our energy in the filtering process when our filters are more likely to lead us to a good purchase.

People have different filters with different weights. Before inspecting more closely, some people weigh the price, some people see how it looks, some people, the professional metascore. I put most weight on the overall user review score. Not everyone is like that, but there are people like that.

There are niche games that don't get the customers they deserve because they are turned away by people downvoting because it was not their niche. With the possibility to see if people feel strongly about the product, people who assess the product to be their niche may still inspect it more closely even with mixed reviews.

[Radene]

You invest some time into figuring out.

[B l u e b e r r y P o p t a r t]

Does it really matter? You shouldn't be going on what reviews say on games if you want to get them anyways. They are biased towards the ones that wrote them and most of them are dumb reviews anyways.

[Hextravert]

You told us exactly what the differences are.

Jagulars님이 먼저 게시:

There are games that get mixed reviews because they are "meh" and people generally don't like or dislike them.

There are also games that get mixed reviews because some players love them and some players hate them.

How can you know the difference?

[Start_Running]

You're asking the wrong question OP.
First understand that all reviews are opinions and there for subject interpretations skewed by personal bias.


The second thing is that whether a review says yes or no, you should read the commentary that comes with it so you can learn why. What did the writer like? what did the writer dislike? Do those same dislikes apply to you?

What someone hates, other people will love.

[Gus the Crocodile]

Standard deviation of...what, exactly? I may just be tired, but I don't see the application.

One game's 47% only meaningfully differs from another game's 47% by sample size, but sample size is already published, there's no need to infer it.

Standard error calculations would be more relevant if this was a system where there were meaningfully different ways for the same sample size to reach the same mean (eg. if reviews were averaged from scores out of 10).

[Jagulars]

It seems I need to explain myself better. There are thousands of games. Few games are worth checking for me. Personally I trust user reviews more than a pretty facade. Consequently, I check only ones that have "overwhelmingly" or "very positive" reviews. This causes me to miss games that I could like because they are targeted to my niche but have "mixed" or "mostly positive" reviews. I don't have time or interest to delve deeper into every game to see if it's worth my time.

[Start_Running]

Jagulars님이 먼저 게시:

It seems I need to explain myself better. There are thousands of games. Few games are worth checking for me. Personally I trust user reviews more than a pretty facade. Consequently, I check only ones that have "overwhelmingly" or "very positive" reviews. This causes me to miss games that I could like because they are targeted to my niche but have "mixed" or "mostly positive" reviews. I don't have time or interest to delve deeper into every game to see if it's worth my time.

Well then you will miss games. that you would like. That's the curse of laziness.
You will only ever find the stuff that is geared towards mass market appeal.

[Quint the Alligator Snapper]

OP's issue is one good reason why a neutral or other middle option would be useful.

Hextravert님이 먼저 게시:

You told us exactly what the differences are. :DEALWITHIT:

Jagulars님이 먼저 게시:

There are games that get mixed reviews because they are "meh" and people generally don't like or dislike them.

There are also games that get mixed reviews because some players love them and some players hate them.

How can you know the difference?

OP's question is how that difference can be detected in short order, rather than trudging through a bunch of reviews just to find out.

[Quint the Alligator Snapper]

Gus the Crocodile님이 먼저 게시:

Standard deviation of...what, exactly? I may just be tired, but I don't see the application.

One game's 47% only meaningfully differs from another game's 47% by sample size, but sample size is already published, there's no need to infer it.

Standard error calculations would be more relevant if this was a system where there were meaningfully different ways for the same sample size to reach the same mean (eg. if reviews were averaged from scores out of 10).

Not standard error (since we're not sampling), but simply standard deviation of intensity of like/dislike.

It's not currently possible to calculate this because the review system is inherently bipolar -- you either upvote or downvote.

But with the addition of a middle option, people with less strong opinions could choose that, and thus a higher ratio of middle option ratings to total ratings would be an indicator of low intensity. That's what the OP is talking about.

[Jagulars]

Start_Running님이 먼저 게시:

Well then you will miss games. that you would like. That's the curse of laziness.
You will only ever find the stuff that is geared towards mass market appeal.

The game publisher, developer and Steam will also lose because I don't buy the game. It's a lose-lose situation which is easy to turn win-win by adding the middle voting option suggested by the people on this thread.

[Start_Running]

Jagulars님이 먼저 게시:

The game publisher, developer and Steam will also lose because I don't buy the game. It's a lose-lose situation which is easy to turn win-win by adding the middle voting option suggested by the people on this thread.

Actually they lose nothing if you don't buy. One of the perks of digital. In physical retail each copy costs money to produce and ship and each non-sale costs them about as much as it did to produce that physical unit. So if you don't buy digital.. well. it's not like it costs them anything.

Again keep in mind you openly admit to being too lazy to actually be bothered to read the reviews to understand the context in which they are given. And you apparently lack the self awareness to make decisions without so... yeah. You're the opnly one being negatively affected here. Though I supposed that also means you get to keep your money and do other fun stuff with it.

[Jagulars]

Start_Running님이 먼저 게시:

Actually they lose nothing if you don't buy. One of the perks of digital. In physical retail each copy costs money to produce and ship and each non-sale costs them about as much as it did to produce that physical unit. So if you don't buy digital.. well. it's not like it costs them anything.

Again keep in mind you openly admit to being too lazy to actually be bothered to read the reviews to understand the context in which they are given. And you apparently lack the self awareness to make decisions without so... yeah. You're the opnly one being negatively affected here. Though I supposed that also means you get to keep your money and do other fun stuff with it.

They lose what the money I would use could be invested in.

Do you expect me to read the reviews of every game published on Steam and call it laziness if I don't?

[Start_Running]

Jagulars님이 먼저 게시:

They lose what the money I would use could be invested in.

No loss on their books that. Only a loss if they spent something. From there perspective, there's no proof that you would have bought it any way. Again a switch brought about by digital. Physical runs are produced based on careful estimations and analysis as to how many people they expect to purchase. Any shortfall in purchases can be costly.

Digital has no such problem. Again. Each unit costs them nothing. So they're no expectations or such in play. Your non-purchase is simply renders you a non-entity.

Do you expect me to read the reviews of every game published on Steam and call it laziness if I don't?

If it's a game you're interested iun purchasing.. I'd expect that'd be basic research. But then I'm opne of those people who don't buy on trends or popularity as you seem to do. I buy based on research.

Again. You're complaining that you're missing games because you're too lazy to research games that are not mass market.

[Jagulars]

That's true. Digital copies cost nothing, and also it's better to do research instead of buying based on public opinion.

Consider that you put your money on your bank account instead of investing it. It will stay there and most people don't consider it as loss. Still, some people prefer to invest that money to gain more, am I right?

And actually, you're right. I'm too lazy to research those games more in-depth. Still I'm offering my money here. Why not take it?

[Start_Running]

Jagulars님이 먼저 게시:

Consider that you put your money on your bank account instead of investing it. It will stay there and most people don't consider it as loss. Still, some people prefer to invest that money to gain more, am I right?

And those people in turn do a lot of research or hire people to do the research for them.

In the end effort always yields greater rewards.