The Tao of Gaming

Boardgames and lesser pursuits

Smoking Sevens

Larry, in an apparently serious move, rates another dozen games from the Gathering. To quote Tom.Hancock:

I figure you were just baiting the haters with this article until I read that last comment. Yikes

To be honest, so did I.

Larry, in his comments:

There are other gamers–Chris Farrell is an outstanding example–who seem to have such high standards that they apparently dislike (or at least find fault with) most of the games they try.

You know, I’ve written an article or two about “Avoiding becoming or sounding like the Anton Ego of game reviewing.” (But I didn’t post it. It sounded pompous). It’s perfectly reasonable to say:

  • “This game has mild novelty value, isn’t bad, but isn’t special.”
  • “I enjoy this despite the flaws.”
  • “There’s a good idea buried here, and I want to explore it a few more times, then I’m done.”

For Larry, it seems that all of these statements will earn a “7” rating from him. He’s trying to map multiple axis of information (innovation, lack of flaws, depth, time/fun ratio, etc) into one degree of freedom.

A publisher wouldn’t (knowingly) publish a game that was universally considered too long, boring, unbalanced and derivative. A fast, enjoyable, deep, innovative & elegant game shows up a few times a decade (if we’re lucky). Most games score high marks in some areas, and low in others. If they are good enough to get published (by a “big” company) odds are the good outweighs the bad, or it’s catering to a niche that focuses on the ‘good’ aspects. FFG’s games are catering to people who don’t mind 3+ hour games with lots of chrome.

I don’t think I go out of my way to play games I’ll hate, but I’ve reported several flawed games that I still played 5+ times. Larry, have you considered just switching to a 1-5 (or four) star system? Three stars can be your seven….

Advertisements

Written by taogaming

April 26, 2009 at 11:48 am

Posted in Convention Reports, Ramblings

Tagged with

21 Responses

Subscribe to comments with RSS.

  1. Jesus, Brian, I expected some flak from some of the nimrods who posted comments, but not friendly fire from someone like you…

    The problem seems to be that I posted the games in order of rating, so that all the 7s were bunched up in one article. But Good God, I did say that was what I was doing. I reported on 52 games that I had played in the past six months. 15 of them got a 7 rating. Is that such an outrageously high percentage? I do focus on games that I think I’ll like and I’m usually successful.

    The key statement in your “Anton Ego” example is “I enjoyed this despite the flaws”. If I enjoy a game, then it will probably get a 7 rating or so. That’s the purpose of a rating, right, to reflect how much the rater likes the game? Sure, it’s one dimensional and possibly misleading; that’s why I provide detailed comments, so other players can judge for themselves if they’ll like the game or not. But I hope a game doesn’t have to be perfect in order for you to enjoy it and be willing to play it again. I hope a single shortcoming that keeps a game from being great but still leaves it fun to play doesn’t consign it to a rating of 5 or less.

    To say that a rating tries to map multiple things into one value is accurate, but irrelevant. Everybody rates games. Hell, I get criticized because I don’t list my ratings on the Geek. So for good or for bad, we want that single number to summarize our feelings for a game.

    I could have given my 7 rating for each of those games and then said for each one, “good game, well worth playing, but not one I’ll often suggest”, which is kind of what a 7 rating means to me. But that obviously would have been useless. Instead, I briefly outlined the plusses and minuses for each design as I saw it. Maybe I overemphasized some of the flaws for the purposes of illustration, but it’s hard to get too nuanced in a short paragraph. Each of those games is one I’ll happily play. Even Finca, which was probably the weakest of the 7s, is a nice little game. If my group wants to try it out (which is very likely), I ‘ll teach it and play it and no doubt enjoy myself. I can see it possibly sliding to a 6 with repeated play, but I don’t think it’ll ever get to the point where it’s a below average design for me. However, that doesn’t mean I can’t wistfully wish the game was even better, particularly given its HiG heritage.

    I’ve gotten so many complaints over this that I did a bit of self-evaluation to see if I was really off base with the articles or my personal rating system. And I honestly don’t see the problem here, other than one, perhaps, of presentation. Most of the games I play have a 7 rating or higher because I plan it that way. There are plenty with ratings of 5 or less (as the third article in the series shows), but I do my best to avoid those games, because who wants to suffer through a bad game?

    Maybe part of this has to do with my love of trying out new games. Even if I wind up playing a game only once, it’s important to me that I do so, as I feel I’m missing out on that joy of discovery if I don’t. So perhaps that sense of discovery adds to the good feelings I get from playing a game for the first time. I’m not at all sure about this theory, because I’ve played plenty of games for the first time that I disliked (as Part 3 shows), but it’s a possible explanation for the raised eyebrows the articles have inspired, when I don’t see anything unusual with my conclusions.

    Larry Levy

    April 26, 2009 at 1:08 pm

  2. For what it’s worth, ‘7’ is also my most common rating. 19% of my ratings are ‘7’, and just under 19% are ‘6’. On a scale with 10 buckets, to put 30% of the items in a single bucket seems like quite a few.

    Eric Brosius

    April 26, 2009 at 1:14 pm

  3. For what it’s worth, ‘7’ is also my most common rating. 19% of my ratings are ‘7’, and just under 19% are ‘6’. On a scale with 10 buckets, to put 30% of the items in a single bucket seems like quite a few.

    Eric Brosius

    April 26, 2009 at 1:14 pm

  4. Fair enough, Eric. It’s a matter of degree, but 28% might be a bit much. My ratings too form a bell curve distribution, with few games at the fringes and a lot of concentration in the 6-7 range. That’s one of the reasons I switched from using strictly integers to adding half points (originally, it was only 6.5 and 7.5), to distinguish all the games in those ranges. I guess I could try to expand things more, but I started out with the reasonable meanings of “pretty good”, “good”, “very good”, and “great” for 6, 7, 8, and 9, and it’s served me well to date. I suspect it’s too late to try something new.

    Larry Levy

    April 26, 2009 at 2:04 pm

  5. The point is that the (numerical) ratings don’t add information, and grouping them so tightly seems (to me as well as others) odd. What’s the point of a ten point scale if it’s only effectively a seven point scale. If you only have 10 numbers, you can’t map all those meanings onto them. And look at your part 1, you rate another 15 games 7.5, then 3 8.0s and 2 8.5s. In part 3 you have 5 6.5s, 4 6s, a 5.5, a 5, four 4s and two threes.

    You are adding points to the scale because most games are roughly average. But it looks like compression of the scale. I understand that its pointless to differentiate between “Atrocious,” “Abysmall” “Bad” and “Terrible”, but if you use a wide scale then use it. If you aren’t going to use a wide scale then using a star system seems reasonable.

    Put scientifically … you are implying a numerical precision (6.5 vs 6). But its a gut feel. If you’d use “Good” “Pretty Good” “Great” people would accept it. Then, by ordering reviews by number you just emphasize that most games are average-good, but ‘7’ to some people is ‘good-good’.

    Personally, perhaps you should just use Steffan’s FUDGE levels (Terrible, Poor, Mediocre, Fair, Good, Great, Superb). If you just search and replaced ‘7’ with Good, 6.5 with Fair and 7.5 with “Great” you’d a) be saying the same thing and b) everyone would understand you clearly.

    By saying “7” you let people read lots into it. And where there’s reading, there is misreading.

    When I read the second part, I had mentally changed all the 7.5s to the previous article to 7s. So my thought was “A second article of average games?”

    This isn’t about friendly fire, it’s about criticism and form. There’s quite a bit of useful information in your paragraphs, but the sevens stick out and wave a red flag.

    Incidentally, I lump nearly 45% of games into 5-6 (on BGG’s numerically rating), but I don’t talk about numbers in my reviews. That makes sense, since a ‘5’ or ‘6’ game is one I’ll play a few times because its not bad, but has nothing particularly gripping.

    The BGG # is a quick and dirty guide. Good for meaningless stats geekery. Attaching them to your reviews is doing yourself a disservice.

    Brian

    April 26, 2009 at 2:29 pm

  6. The point is that the (numerical) ratings don’t add information, and grouping them so tightly seems (to me as well as others) odd. What’s the point of a ten point scale if it’s only effectively a seven point scale. If you only have 10 numbers, you can’t map all those meanings onto them. And look at your part 1, you rate another 15 games 7.5, then 3 8.0s and 2 8.5s. In part 3 you have 5 6.5s, 4 6s, a 5.5, a 5, four 4s and two threes.

    You are adding points to the scale because most games are roughly average. But it looks like compression of the scale. I understand that its pointless to differentiate between “Atrocious,” “Abysmall” “Bad” and “Terrible”, but if you use a wide scale then use it. If you aren’t going to use a wide scale then using a star system seems reasonable.

    Put scientifically … you are implying a numerical precision (6.5 vs 6). But its a gut feel. If you’d use “Good” “Pretty Good” “Great” people would accept it. Then, by ordering reviews by number you just emphasize that most games are average-good, but ‘7’ to some people is ‘good-good’.

    Personally, perhaps you should just use Steffan’s FUDGE levels (Terrible, Poor, Mediocre, Fair, Good, Great, Superb). If you just search and replaced ‘7’ with Good, 6.5 with Fair and 7.5 with “Great” you’d a) be saying the same thing and b) everyone would understand you clearly.

    By saying “7” you let people read lots into it. And where there’s reading, there is misreading.

    When I read the second part, I had mentally changed all the 7.5s to the previous article to 7s. So my thought was “A second article of average games?”

    This isn’t about friendly fire, it’s about criticism and form. There’s quite a bit of useful information in your paragraphs, but the sevens stick out and wave a red flag.

    Incidentally, I lump nearly 45% of games into 5-6 (on BGG’s numerically rating), but I don’t talk about numbers in my reviews. That makes sense, since a ‘5’ or ‘6’ game is one I’ll play a few times because its not bad, but has nothing particularly gripping.

    The BGG # is a quick and dirty guide. Good for meaningless stats geekery. Attaching them to your reviews is doing yourself a disservice.

    Brian

    April 26, 2009 at 2:29 pm

  7. The BGG # is a quick and dirty guide. Good for meaningless stats geekery. Attaching them to your reviews is doing yourself a disservice.

    I guess it is. And I usually don’t put a numerical rating in my standard reviews, but I got into the habit of doing so for my Gathering reports, maybe because there was less information there. But in retrospect, it was a mistake and one I won’t be repeating.

    I’m actually glad to see that you concentrate so many of your ratings into 2 points of the scale, because it indicates that my concentration in the 6-7 range isn’t that bizarre. I had figured there was no harm in giving these ratings, on the assumption that information is always good. But your point about them being misleading and overly elevated in importance is valid, so I’ll be following your lead and suppressing them in the future.

    I don’t think I need to alter my rating scheme, as it works very well for me. I see no need to say that just because the rating points are there, I have to use them all the time. I have only five games I rate a 10, which to me, gives that rating meaning. That’s just a philosophical difference in how people do things, so there’s no right or wrong. But I’m comfortable with what I have, so just keeping the ratings to myself should work fine in the future.

    I do appreciate the comments, Brian. I admit that the firestorm over the articles frustrated me a bit, particularly since some of the criticism was a bit more personal than yours. I normally don’t mind the jibes, but they seemed so concentrated on what I felt was a small detail that it had me scratching my head a bit. But there’s truth in what you and some of the others have said, so I’ll make the necessary adjustments.

    Larry Levy

    April 26, 2009 at 3:19 pm

  8. The BGG # is a quick and dirty guide. Good for meaningless stats geekery. Attaching them to your reviews is doing yourself a disservice.

    I guess it is. And I usually don’t put a numerical rating in my standard reviews, but I got into the habit of doing so for my Gathering reports, maybe because there was less information there. But in retrospect, it was a mistake and one I won’t be repeating.

    I’m actually glad to see that you concentrate so many of your ratings into 2 points of the scale, because it indicates that my concentration in the 6-7 range isn’t that bizarre. I had figured there was no harm in giving these ratings, on the assumption that information is always good. But your point about them being misleading and overly elevated in importance is valid, so I’ll be following your lead and suppressing them in the future.

    I don’t think I need to alter my rating scheme, as it works very well for me. I see no need to say that just because the rating points are there, I have to use them all the time. I have only five games I rate a 10, which to me, gives that rating meaning. That’s just a philosophical difference in how people do things, so there’s no right or wrong. But I’m comfortable with what I have, so just keeping the ratings to myself should work fine in the future.

    I do appreciate the comments, Brian. I admit that the firestorm over the articles frustrated me a bit, particularly since some of the criticism was a bit more personal than yours. I normally don’t mind the jibes, but they seemed so concentrated on what I felt was a small detail that it had me scratching my head a bit. But there’s truth in what you and some of the others have said, so I’ll make the necessary adjustments.

    Larry Levy

    April 26, 2009 at 3:19 pm

  9. Larry, it may seem like a firestorm, but I don’t think people are meaning to dogpile on you. I’d view it as a compliment—namely, people respect your opinions, and so they feel like it’s worth while to respond in detail to your post so they can explore this issue of how people value different games.

    I certainly read all three parts of the write-up with interest. I agree with your opinions more than I agree with the opinions of the average person, though (obviously) there are some instances in which we differ markedly.

    Eric Brosius

    April 26, 2009 at 5:42 pm

  10. Larry’s “problem” is easy to understand: there are too many–WAY too many–good but not great games on the market. And they are good: they work, they do what they’re designed to do, there are no huge flaws, etc. But they don’t stand out from the pack. So I’m not surprised he tried to differentiate between that mass in the middle with some extra qualifiers (weak, strong, etc.). I mean, let’s say you have this set of games, A-J, and here’s your ranking:

    10: A
    9:
    8:
    7: BCDEFGH
    6:
    5:
    4:
    3: I
    2: J
    1:

    Does using a 3-star system add value?

    ***: A
    **: BCDEFGH
    *: IJ

    What about an ordinal ranking?

    ABCDEFGHIJ

    I think the latter two imply a closer relationship in quality where there shouldn’t be one. A is vastly better than B. H is vastly better than I. Compressing the scale loses that distinction.

    So Larry, don’t give in to these nattering nabobs. Fight for your 7s!

    Jon Waddington

    April 27, 2009 at 10:29 am

  11. Larry’s “problem” is easy to understand: there are too many–WAY too many–good but not great games on the market. And they are good: they work, they do what they’re designed to do, there are no huge flaws, etc. But they don’t stand out from the pack. So I’m not surprised he tried to differentiate between that mass in the middle with some extra qualifiers (weak, strong, etc.). I mean, let’s say you have this set of games, A-J, and here’s your ranking:

    10: A
    9:
    8:
    7: BCDEFGH
    6:
    5:
    4:
    3: I
    2: J
    1:

    Does using a 3-star system add value?

    ***: A
    **: BCDEFGH
    *: IJ

    What about an ordinal ranking?

    ABCDEFGHIJ

    I think the latter two imply a closer relationship in quality where there shouldn’t be one. A is vastly better than B. H is vastly better than I. Compressing the scale loses that distinction.

    So Larry, don’t give in to these nattering nabobs. Fight for your 7s!

    Jon Waddington

    April 27, 2009 at 10:29 am

  12. Larry’s “problem” is easy to understand: there are too many–WAY too many–good but not great games on the market. And they are good: they work, they do what they’re designed to do, there are no huge flaws, etc. But they don’t stand out from the pack. So I’m not surprised he tried to differentiate between that mass in the middle with some extra qualifiers (weak, strong, etc.). I mean, let’s say you have this set of games, A-J, and here’s your ranking:

    10: A
    9:
    8:
    7: BCDEFGH
    6:
    5:
    4:
    3: I
    2: J
    1:

    Does using a 3-star system add value?

    ***: A
    **: BCDEFGH
    *: IJ

    What about an ordinal ranking?

    ABCDEFGHIJ

    I think the latter two imply a closer relationship in quality where there shouldn’t be one. A is vastly better than B. H is vastly better than I. Compressing the scale loses that distinction.

    So Larry, don’t give in to these nattering nabobs. Fight for your 7s!

    Jon Waddington

    April 27, 2009 at 10:29 am

  13. Oh, I should add, I think all Larry’s .5 modifiers are his alternate way of writing “strong” or “weak”. He’s just trying to dilate the 6-7-8 area to glean more useful differentiation, which is hard when they’re all basically the same. Changing the number of points (20,10,7,5,3) doesn’t really help with that, IMHO, or add more useful information. Dogpiling is still gonna happen, unless he goes completely Anton.

    Jon Waddington

    April 27, 2009 at 10:44 am

  14. 10: A
    9:
    8:
    7: BCDEFGH
    6:
    5:
    4:
    3: I
    2: J
    1:

    To be fair, that’s a strawman. Larry really showed
    10:
    9:
    8: AB
    7: CDEFG
    6: HI
    5:
    4: J
    3:
    2:
    1:

    (I’m using ordinals, but that’s basically correct). If Larry had used your example, I think he’d have gotten less flak, because the 3 point jump between A and the rest is huge. But his highest rating was 1.5. He probably meant it the same, to me 1.5 is “One Standard Deviation” and to Larry it’s “Two Standard Deviations.” (or 1.5). Everyone reads their own precision into numbers.

    And I agree that he’ll have issues no matter what, since he’s trying to map lots of data into one point. (I’m tempted to invoke Tufte here). Hence my recommendation that he embrace ambiguous language and denounce arithmetic precision.

    Brian

    April 27, 2009 at 11:10 am

  15. 10: A
    9:
    8:
    7: BCDEFGH
    6:
    5:
    4:
    3: I
    2: J
    1:

    To be fair, that’s a strawman. Larry really showed
    10:
    9:
    8: AB
    7: CDEFG
    6: HI
    5:
    4: J
    3:
    2:
    1:

    (I’m using ordinals, but that’s basically correct). If Larry had used your example, I think he’d have gotten less flak, because the 3 point jump between A and the rest is huge. But his highest rating was 1.5. He probably meant it the same, to me 1.5 is “One Standard Deviation” and to Larry it’s “Two Standard Deviations.” (or 1.5). Everyone reads their own precision into numbers.

    And I agree that he’ll have issues no matter what, since he’s trying to map lots of data into one point. (I’m tempted to invoke Tufte here). Hence my recommendation that he embrace ambiguous language and denounce arithmetic precision.

    Brian

    April 27, 2009 at 11:10 am

  16. 10: A
    9:
    8:
    7: BCDEFGH
    6:
    5:
    4:
    3: I
    2: J
    1:

    To be fair, that’s a strawman. Larry really showed
    10:
    9:
    8: AB
    7: CDEFG
    6: HI
    5:
    4: J
    3:
    2:
    1:

    (I’m using ordinals, but that’s basically correct). If Larry had used your example, I think he’d have gotten less flak, because the 3 point jump between A and the rest is huge. But his highest rating was 1.5. He probably meant it the same, to me 1.5 is “One Standard Deviation” and to Larry it’s “Two Standard Deviations.” (or 1.5). Everyone reads their own precision into numbers.

    And I agree that he’ll have issues no matter what, since he’s trying to map lots of data into one point. (I’m tempted to invoke Tufte here). Hence my recommendation that he embrace ambiguous language and denounce arithmetic precision.

    Brian

    April 27, 2009 at 11:10 am

  17. Well, dammit, it was my strawman, so stop picking on him.

    Actually, it does map pretty closely to how I feel about most games, most years.

    Jon Waddington

    April 27, 2009 at 11:24 am

  18. Jon, this whole episode was worth it just so that I could hear someone use the phrase “nattering nabobs” again! Thanks!

    And Brian, in his last comment, is correct. The 1.5 ratings point gap between an 8.5 and a 7 is enormous in my eyes. That’s because you’re approaching the end of the curve. But in some people’s systems, it’s no big deal. So the lesson is, use the words and keep the numbers to yourself, because others are liable to convert them to their own system and therefore misinterpret them.

    Larry Levy

    April 27, 2009 at 12:32 pm

  19. My ratings tend to be like:

    10: Best game of a 1-2 year period
    9: ~2 Best games of the year.
    8: The rest of the very good games that come out.
    7: All the above average, decent games that are good, but not as good as the great games so I will only play them a couple times. Say, 60th to 80th percentile of released games for the year
    6: Average games, stuff that is basically like a different game that I have played already.
    5-: Below average, bad, terrible…

    Alexfrog

    April 28, 2009 at 3:51 pm

  20. My ratings tend to be like:

    10: Best game of a 1-2 year period
    9: ~2 Best games of the year.
    8: The rest of the very good games that come out.
    7: All the above average, decent games that are good, but not as good as the great games so I will only play them a couple times. Say, 60th to 80th percentile of released games for the year
    6: Average games, stuff that is basically like a different game that I have played already.
    5-: Below average, bad, terrible…

    Alexfrog

    April 28, 2009 at 3:51 pm

  21. My ratings tend to be like:

    10: Best game of a 1-2 year period
    9: ~2 Best games of the year.
    8: The rest of the very good games that come out.
    7: All the above average, decent games that are good, but not as good as the great games so I will only play them a couple times. Say, 60th to 80th percentile of released games for the year
    6: Average games, stuff that is basically like a different game that I have played already.
    5-: Below average, bad, terrible…

    Alexfrog

    April 28, 2009 at 3:51 pm


Comments are closed.

%d bloggers like this: