I'm just curious about how people view game ratings, and how they rate games on BGG. Rather than get opinion, I figured I'd get some data. I'm sure people will think of clever follow on questions...



EDIT -- Added a follow on poll for high average people.

very interesting!
thanks for the articulated poll.

Brian, I assume the top choice in Poll 6 should read "Always" instead of "Never"?
If not, I would like to know the difference between the two 'nevers'

Yes, it should be always.

I think my average is above 5.5 because I really only have one game on the "1" rating, so my scale really averages closer to 6.5

My average is very close to 5.5, yet my most common vote is 7. The 7's are offset by the fact that I have more 1's than 10's and a lot more 2's than 9's.

I match 1s and 10s (and 2/9, 4/7, 5/6) closely, but I have twice as many "3"s than 8s. I could probably go and find more 1s and 2s, but what's the point?

3-5 is just the range of games I'll never suggest, and just shows how grudgingly I'll play it. So that keeps me at 5.35. I suspect that lots of people have a fair number of games in that range, but have plenty of games they've played once long before joining, hate, but haven't rated.

Even so, I'm still surprised that only 20% of the people responding have averages less than 6!

That can mean one of two things---more likely it's a combination of both:

- ratings are inflated, meaning that on average people tend to think of an average title as one they'd award with a 7 instead of a 5.5---rather, people are very reluctant to hand out low numbers like 1 and 2;
- people only play what they like, and then adhere to the interpretation of the scale to rate the game, causing the average rate to rise as well.

Of course the ratings should skew high. When you buy a game, aren't you trying to buy a game that you'll like, one that you'll usually want to play? I sure as hell am.

If I buy a game and it's only a "Meh, will play sporadically" 6, then that's a failure, likely a waste of money.

I'll try to compile and post actual cross-BGG data against these questions. It'll be interesting to see how well self-reporting lines up with reality.

I rate games I don't buy (and the ratings for those games is significantly lower than for the games I buy.) I often play a game at a friend's request, even though I don't particularly expect to enjoy it. Usually my suspicions are correct.

I also suspect the average game rater's average rating is higher than the average rating, because people who rate a lot of games probably rate lower (a cynic might say we're "jaded".) I'm not at all surprised that most people have high averages, but I'll bet this is driven by people who rate relatively few games.

Yes this is true. There's a very strong correlation between number of games rated (more) leading to lower average ratings (less).

This could be partly (a) people who rate more games are trying more of everything (b) there's only a fraction of games which you personally would qualify for a 9 or 10 rating and people tend to find those earlier rather than later, being sensible folk in general and (c) just in general people start at the "better" or higher rated games (games that their buddies recommend) and work their way deeper over time, tending to therefore be exposed earlier to games which are more generally well received.

"Jaded" doesn't need to enter into it, as such, but it could be that too.

If I could find a way to consistently and accurately figure out which games I will rate a 7 or higher, and only buy those, I certainly would. And I don't buy games which I have played that I think will end up with a 6 rating, with very rare exception.

But I definitely enjoy trying new games, and for many games the only way I'm likely to get to try them is by picking up a copy, often with very little useful information to guide my purchase. Fortunately, I've found that most games can be traded or sold for reasonable value, largely limited the "waste of money" factor.

Joe (who has an avergae rating of 5.43 in his ratings but 5.6 on BGG because here Settlers, Settlers 3-D, Travel Settlers, etc. each are separate entries, where I count them as equivalent, among other examples)

Ditto. My average for my collection is slightly over 7. (I don't necessarily get rid of games I rate a six or lower, but those tend to get slowly purged). But sometimes I take a chance and buy a game that may be an 8+ and wind up with a lower score. Or my tastes change...

My ratings used to be much lower, but now I'm introducing selection bias by not playing things I probably won't enjoy (Munchkin part 27). This has resulted in a higher average of late.

I have to say my ratings average is a little higher than some of you. I think this is for 2 reasons.

1. I think I am a pretty good judge of what I like over all and tend to play those games.

2. Probably more importantly, I don't rate games after a single play. So there are games with potentialy low ratings from me out there that I tend to avoid. Before you go all medieval on me for this I don't think 1 play is a fair charecterization of a game.
Case in point; last night I played Warriors of God. The dice were against me the whole game. I even wanted to call the game when it reached a point that seemed beyond recovery. We went through 1 more turn at the behest of my opponent when this became obvious to him as well. Does this mean this is a bad game? Parts of it were less than enjoyable, but who really enjoys a string of bad luck? I wouldn't want to rate this game after that 1 play. So I withhold my rating until I can play again. Maybe that will be never, maybe next week. I would rather leave the game unrated than let a less than optimal game experience cloud my judgement. I want to play more than once so I know that it is the game that deserves my rating and not something else like luck, most likely, in this case. I would rather withhold judgement than condem a game were I may not be able to identify some external stimulus that is driving down my rating.

Awwww.

This was bound to come up again here, now wasn't it. And we're all just waiting for the cue to have our fun. And you want to deprive us. That's just not playing fair!

(So I'm not going to rate this thread until I get one more post in on it.)

Ok here's the real response.

You're right. Your first play may be skewed up or down as an experience compared to what you might have got from repeat plays. And presumably any negative (under rating) first experiences are less likely to result in repeat and (ratings corrective) experiences than the skews upwards.

But you know what? It's still meaningful data. If 100 people play the game and 10 are turned away by a negative but off-centred first experience, that's still meaningful information to include in the fuzzy "one number describes the overall appeal" game ratings system.

Not only that, but if your negative experience is square centre with what you would expect from repeat plays, we really needed to know about it.

In either case, it's a useful and meaningful contribution to the game average.

Furthermore, the most useful information we could have in trying to set up a personalised recommendation system would be what people have tried that they didn't like. Your dislikes are an essential part of aligning your preferences with other people's enough to help us use your experiences to make recommendations to them. Your negative experiences would help us make more meaningful recommendations to other gamers.

Ratings on 1st play? Bring 'em on!

Yeah, I always rate after my first play. My tendencies are to not rate extremely high until I'm sure it deserves that high of a rating over time, but I have no problems giving out a 7 or 8 after a first play if it's something I know I'm going to like. If it's something I like, surely I'll play it again and revise my rating if I feel the need.

However, I have no problem giving a low rating after one play. I don't need to, nor do I want to, play a game 5 times to know that I never want to play it again. There are so many games out there, it would be a waste to sit around playing stuff I don't like repeatedly. In fact, almost all the games I've rated 1-3 are probably ones I've only played once. I feel like I've played enough games by now where I know if a game just isn't my thing.

If I played whatever random games are available, my average rating would be lower than it is. But I consciously choose to play games that look interesting to me and to avoid playing games that don't look interesting to me.

It's just like how most movies are pretty lame, but I enjoy most movies I actually see, because I read reviews, talk with friends who've seen them, and use common sense when deciding what films I want to see.

Great poll!

It had me look at my ratings which really wasn't a surprise. I'm at 6.03. When I give a six, to me, that is average. I like it okay, will play but I'm not clamoring for it to get to the table. A five for me is one I don't care for much but would still play. A ten is a true classic for me (whether I play it anymore or not) and will always love. A ten takes some time and repeated play for me to rate it at that level.

I also rate after playing the game once but I do revisit my ratings on a regular basis. My average has started trending upward over the last year or so. It was at 5.92 at one point. I feel I am playing some better games and have a regular game group who gives some of those a shot with me. I'm also not buying as willy-nilly as I had before.

If we all did that, there would be no use for the lower ratings. Why on Earth would people torture themselves by playing a game they don't enjoy again and again?

My ratings are above average because I actively seek out the games I think I would like, and try to avoid the games I don't. The few games I have rated low are mostly suggested by people I enjoy playing with, and thus tried against my better judgement.

Interesting poll, but it didn't say anything new about me. I'm not that surprised by the fact that a lot of people have the same practice as I.

When I started rating on BGG, I quickly found out that '7' is an average game. Whose fault is that? Well, the one who wrote the rating guidelines. A '5' which is "Slightly boring", now that's a BAD game in my opinion. '1' clearly broken - defies the definition of a game, isn't really a rating, it's either 1) just a childish exaggeration that "I don't like this game"
or 2) an assuption that we're to include flower pots in the ratings too. No game, no matter how bad it is, can clearly defy the definition of 'game', and any sensible person will realise that. Oh well, we can start an argument about the definition of the word 'game' but... Even the stereotypical "roll a d6, the one who rolls highest wins" is still a game even if it's stupid. Sure, one can sit and fabricate a ruleset for a 'game that isn't a game', but the point isn't that it's theoretically possible, the point is that those kind of games just don't really exist in any shop.

The low ratings just look like someone had fun inventing different exaggerative wordings for crap games. Personally, I don't care how crap a user thinks the game game is. If they think it's crap, that's it and that's all I need to know.

So, I realised I either had to ignore the guidelines, and just rate 1-10 by my own definitions, or set 'average game' as a 7. I chose the latter. For a 'bad' game, I realised 5 was the best definition, and for a total crap game, I realised 3 was the best. So I left out 1, 2, 4 and 6 and rated away.

Also, I rate after first play, mostly. In rare cases I wait till further plays. I'm old and experienced enough gamer that I fell I am perfectly able to judge on one play whether a game is to my tastes, not to my tastes, shallow, deep, having more that I can't judge yet, or whatever. To assume that everyone needs several plays for every game they play in order to judge it is just silly.