« Brakhage | Main | Bryant's Consumer Guide »

April 03, 2024

A FEW WORDS ABOUT LETTER GRADES

Mike D'Angelo writes:

I can't be the only one curious to know why you've decided
to abandon grades after all these years...

Maybe it's not apparent from my Web site — which first graded on the venerable star system but switched early on to letter grades, which appealed because they felt, I don't know, snottier — but I've always had a philosophical problem with grading movies, or handing out stars or hoisting and lowering thumbs, or dealing out movie tickets, or dispensing whatever virtual tchotchkes it is that those crazy Internet kids are dispensing these days.

The common hardline stance is that film criticism is too serious a business to boil down to a consumer-guide rating. "If you, fickle reader, want to know what I think about a movie, goddamnit, you can read my review, where I'm hoping to communicate something a little more complicated about my experience of the film than just, you know, B+, you shallow bastard." That may or may not be bullshit, depending on what you think of my writing and whether you ever use my letter grades as a kind of consumer guide.

But there are practical considerations, too. For one thing, there's the problem of reconciling the graded list, like a bank statement, at year's end. Once Gangs of New York, say, is listed as an A- and Visitor Q is listed as a B, the creation of, say, a top 10 list becomes a fairly rote exercise in skimming the cream off the top of a list of films ordered by letter grades. And that's an exercise that doesn't reflect the ways that movies can gain power and context as the months fly by.

And there's also the little question of accuracy. When I look at my sorted-by-grade list of 2002 films, it's generally a good reflection of my year at the movies. But on a film by film basis, I have reservations. Should Resident Evil, which features zombies and Milla Jovovich, both of which elements continue to entertain on repeat viewings, get bumped out of its C+ ghetto, where it languishes betwixt and between a workmanlike snoozer like Wendigo and a moralizing snoozer like Auto Focus? Shouldn't Storytelling, which I revisited in part on DVD and consequently have no desire ever to experience again, get docked a notch or two rather than being allowed to pal around with the altogether more sophisticated Undercover Brother?

And how much time should I spend rearranging my list? Should I do it at the end of the year, or revise it every time I return from the local googolplex? Do I owe it to my readers to track the changes over time? Do I owe it to the films to make a point of revisiting them, just in case my original opinions change?

Mike wrote me once (I remember all of Mike's email) to express gratitude that I attached letter grades to my reviews, because he never would have been able to decode from the grudgingly admiring review I gave Topsy Turvy the A- that perched atop it. And you know what? Topsy Turvy is a fine film, and a perfectly admirable film. But it didn't deserve the A- I gave it, which was an indicator of, I guess, my level of respect for the picture rather than my active engagement with and/or enjoyment of it. My review is clearly a B+. Maybe even a B.

Here's another question: what the fuck does "B+" mean, anyway? How I've always seen it is B+ is the cut-off point. When the lights go down, that's my baseline expectation. It's what I'm hoping for. If the lights come up and I feel good and satisfied and glad I invested $10 and two hours (often plus travel time from the suburbs to Manhattan, which can be a bitch) in the film, it's at least a B+, with a higher grade possible depending on the film's ambitions and accomplishments. An A- is a movie that I fairly well love. And an A is something I give out altogether too rarely. It's so hard for me to figure out a way to triangulate that reaction when I'm still close enough to a movie to be writing about it. Kundun, for instance, got the grade because my first viewing of it was so dazzling. I remember stumbling across Broadway to Ollie's Noodle Shop, my chest so tight that I could hardly breathe, when it was over. Magnificent, yes. But it's an experience that can't be duplicated. I still loved the film on a second viewing, but I didn't love the film in the same way. I guess it would be impossible. And on DVD, it is diminished further still. Is it no longer an A? It wouldn't seem fair to retroactively knock the movie down a peg or two because it doesn't provide a similarly overwhelming experience on repeat viewings — in most cases, a movie only has to work once, and besides critics almost always have to write about a film based on that single viewing. Second-guessing your own emotional reactions is no way to spend your time.

But I note that in 1997, the year of Kundun, I also handed out As to The Sweet Hereafter and L.A. Confidential. That seems reasonable to me, although I wonder, if I enjoyed L.A. Confidential so damn much, how come I've never had the urge to slip in the DVD that's been sitting on my shelf for five years? Is it my latent fear of Kim Basinger? (Reading my review today, I notice that I had kind words for her performance and therefore wonder what the fuck I was smoking.) But then is it really true that no movie I saw in 2002 — or 2001 for that matter — deserves a flat-out A grade? Nah. Looking back on 2001, I see that I should have handed out As, flat out, to Audition and Mulholland Dr. at a bare minimum. And what's up with Battle Royale, a movie I love as much as anything I've seen in the last five years, languishing in the Bs?

And here's the thing. I had zero enthusiasm — zip, nichts, nada — for constructing a top 10 list for the year 2001, and again for 2002. Keeping running tabs on my opinions, putting the films in boxes as I saw them, sapped my enthusiasm. I could always mix it up as I order and re-order the pictures at the end of the year — but self-consciously, feeling that, if I pulled one film out of the basement and sent another one tumbling down the stairs, I'd have to justify the repudiation of my original reaction. What I find is it's not what I write about the film that's the problem — some of my old reviews make me wince, but they are consistently true documents showing how I reacted to the films they consider. It's the letter grades that often seem untrustworthy. Misleading, even.

Sure, I could continue to affix letter grades to every film, and then just throw them out at the end of the year when it comes time to select the "best." Further, it might shed some light on my own habits to see which films rise and fall in my estimation before the year's end. But if I'm to make the concession that letter grades are often merely placeholders to indicate a snap judgement on a film before time has its way with me, and with said film, is it fair to place them at all? And is it necessary?

I send out an email when new movie reviews get posted. When I wrote about Irreversible, a film which resisted my efforts to grade it according to my usual criteria, I noted that it seems complex reactions to difficult films are unhelpfully reduced by letter grades: "I've elided a letter grade because I'm toying with the idea of no longer assigning grades to movies, a practice that always seemed a little arbitrary. Also, I don't have any idea what the hell I'd do with [Irreversible] anyway, which gets an A+ for technique and sheer impact, something in the B range for substance, and a failing grade for running around the schoolroom with its wee-wee hanging out of its pants and terrorizing the smaller kids. My remarks here are glib, but the film really is something else — I wrestled with it for three days and came up with the review linked above."

If I feel this ambivalent about letter grades months, even years, after a film is released, you can imagine my reaction when I hear, from md'a, that he knows what grade a movie is going to get as soon as the credits roll. That's an admirable quality, and Mike obviously gets a lot of things right. If you know his work, the grades actually are a pretty helpful metric. I wish I were as good at it as he seems to be.

But there is a tendency on the part of film buffs, especially self-published Web-centric film writers, to spend time creating lists that chronicle and compartmentalize their obsessions into something manageable. Mike's Web site itself has influenced a number of like-minded cinephiles who follow his lead of keeping online screening diaries, and have similarly managed to make something worthwhile out of this dedication. I read them, of course. I tip my hat to 'em. But the granularity with which some of them break down their cinephilia, frankly, frightens me. One mailing list I belong to features a weekly compilation of Variety-style "Crix pix," breaking down the reactions of subscribers along the lines of pro, con, and mixed. When the idea was floated of adding more levels of distinction to the system, essentially allowing for VERY pro and VERY con reactions, I objected meekly. Were it up to me, I'd reduce the available reactions to just pro and con, forcing people off the fence entirely. Of course, the suggestion passed overwhelmingly among the members. (I seem to recall somebody lobbying for breaking mixed up into mixed-verging-on-pro and mixed-leaning-backward.) The system still works very well for encapsulating the reaction of a controlled group of moviewatchers, but I still find myself wishing everyone who voted "mixed" would come down on one side or another. The ambiguity they felt in making that decision could then become the indirect subject of their longform writing on the films themselves.

So when Mike D-to-the-A himself recently changed over from the relatively sane A-F scale to a 100 point grading system, you can imagine, I started to itch. (Plus he gave Dawn of the Dead only a 77, which, I don't know, does he mean that 23 percent of the films ever released are better than Romero's masterpiece or what, exactly?)

Anyway. This is a big to-do about very little, actually. I'm planning to move forward without the agita of letter grades. I'm hoping it will spur me to write about more movies, rather than just slapping them on a "films viewed" page with a grade next to them, but we'll see how that turns out. My full-time job has been keeping me really busy lately, so it's hard to work up the consistent enthusiasm for seeing and writing about movies, but I still do love to see 'em and write about 'em. We'll see how it goes. Go ahead and fill up the comments field below if you feel strongly about any of this one way or another, you poor sucker.

-bf-

Posted by Bryant at April 3, 2024 10:26 PM

Trackback Pings

TrackBack URL for this entry:
http://www.deep-focus.com/pcgi-bin/MT-2.21/mt-tb.cgi/9

Comments

for practical reasons grades we´re always comfortable to fall back to. but as a matter of principle it makes sense to give them up, after all art is (rather should but let´s leave it at that) more complicated than to be evaluated by a simple point system, which most of the time takes too much emphasis on its own. hope my school english was enough to make myself clear here.
i´m fine with whatever you do, as long as you keep the reviews coming :D

Posted by: Heiko Stullich at April 4, 2024 03:08 PM

You should never have moved away from the four-star scale, Bryant. It's imprecise enough to refute claims of exact pigeonholing -- à la the 100-point scale, or even the letter-grade system -- but it still provides for a quick, easy indicator of what you think of a particular film.

I can understand the frustration a more vague system like the four-star scale might cause, resulting in a move to something more complex and exacting, and then finally ending with the whole shebang folding back in on itself (or something).

Still, at the end of the day, the review's the thing. Down with grades. Etc.

I give this post a piffling D-

Posted by: John Harrison at April 4, 2024 06:01 PM

Thanks, guys, for the responses. I was thinking about this on the way to work today, and less-complicated systems began appealing to me. I like the way Jonathan Rosenbaum's star ratings, for instance, offer clues ("masterpiece," "must see," "has redeeming facet," etc.) that help you navigate his various digressions. And I think just dropping the pluses and minuses from the letter grade system might go a long way (in other words, movies would be graded A, B, C, D or F without the hedge that the modifiers provide).

Still thinking on this one. For the record, I'd probably give The Hunted a B, Häxan an A, Stone Reader a B, and Irreversible — well, shit, I'm still not sure what I'd give Irreversible. Probably a B, although that doesn't begin to reflect the way I really felt about it.

Posted by: Bryant at April 4, 2024 06:19 PM

Film buffs are, by their very nature, an anal-retentive lot, and so the most complex, exacting method of adjudicating a film will undoubtedly have certain alluring qualities. (Although, for the life of me, I don't understand how anyone could make effect use of the 100-point scale. I mean, for crying out loud, what's the difference between a rating of 76 and one of 77? And surely, unless one is of a stoic disposition, not to mention thoroughly dogmatic, wouldn't the ratings constantly fluctuate, driving one to the brink of insanity and beyond? It doesn't bear thinking about, frankly. D'Angelo, you are a braver man than I.)

At the end of the day, though, handing over a grade isn't an exact science, and so applying it as such seems ridiculous and, to some extent, even gaudy. Something like the four-star scale, however -- or even Rosenbaum's slightly reworked version -- is sufficient to help people navigate their way through your review, or just give them a quick idea of what you thought of a particular film should they wish to save reading what you've written until after they've seen it.

As far as I'm concerned, unless one's writing is obvious, easily digested and composed in broad strokes, appending a simple rating/guide would help your readers avoid confusion and allow them to fully appreciate where you're coming from -- or maybe that's just a lazy way of looking at it. There's no harm, though, in making use of a support (albeit a largely unnecessary one) to prop your opinions up against.

Posted by: John Harrison at April 5, 2024 04:16 AM

Bryant:

As usual, I find your eloquent thoughts very ... thought-provoking. I agree that rating
systems have many problems, and I also believe that the more resolution a rating system has,
the more ridiculous it is. I can't imagine ranking movies on a scale of 100, and your use
of letter grades with pluses and minuses increased your levels from 4 (when you used the
4-star system) to well above that (13 at least?)

But another problem critics do with rating systems is to totally waste the bottom half
of the scale by using that part, wastefully, to indicate degrees of "not recommended" as if
folks really care how un-recommended a film is. And because they waste the bottom half of
the scale, they don't have enough positive ratings left, so they resort to using half-stars!

For instance, many critics use a 4-star scale where they define ratings of 2 or lower as "not
recommended." Being left with only a 3 and a 4, they resort to introducing 1/2 stars so that
more positive ratings are possible. On their scale 0, 1/2, 1, 1-1/2, and 2 all indicate a thumbs
down. So 2 means "Not Recommended" while 1-1/2 means "Really Not Recommended" while 1 means
"Really Really Not Recommended", etc. It's silly, especially since you're still left with 1/2
and 0! By the critics admission I should avoid them all since they all suck. I don't really
care whether it's rated a 1/2 versus a 1 versus a 1-1/2.

This is why I used (when I had time to write short reviews) a 4-star system where *every* one of
the ratings was a recommendation. The more stars, the more I recommended the film. If the movie
got ZERO stars, I didn't recommend it. This way I didn't waste any part of the scale.

**** Superb: A film of the highest interest; a must-see film
*** Very Good: A film of high interest that stands up to multiple viewings
** Good: An interesting film worth a look
* O. K.: A film of some interest, but no great loss if you miss it
0 stars: A film of no interest; skip it

If you ever decide to 'go back', you may consider using some variation of this, but keeping
the levels to four or less is probably best. Of course NO levels (no ratings) is in some
ways even better.

With that, keep up the great work my friend.

Dan

Posted by: Dan Balogh at April 6, 2024 12:37 AM

Bryant:

A nice survey of the problems involved with movie grading systems. I'm glad D'Angelo's 100-point scale works for him (and it seems obvious to me that it does). Insane-making, says I.

It seems to me that the next frontier in movie rating systems is for someone to rank every film they've seen from 1 to 3624 (or whatever), then figure out the rank of every new film seen in comparison to that list and readjust the whole accordingly. ("'Irreversible' is the 345th best film I've ever seen, so it's a 345 out of 4211 as of today!")

The more film commentary and criticism I read, the less interesting letter or numerical grades become. It's hard to decide what they should mean: are they recommendations, or are they reflective of your opinion of the movie? If the former, they rarely work; when people ask me if I'd recommend a movie or not, I tend to ask if it interests them. If so, I recommend it, regardless of what I thought. And you've thoroughly discussed the problems of coming up with a grading system that accurately reflects "what you thought" of a movie.

Venturing much beyond thumbs up/thumbs down seems to me to muddy things, rather than clarify them.

James

Posted by: James Callan at April 14, 2024 02:45 PM

Two things:

(1) Grades/ratings/whatever are not "interesting" per se. They're shorthand, and extremely useful to those of us who don't want to read (or even skim) a review until we've seen the film in question. Example: I'm still on the fence about seeing THE HUNTED. If Bryant thought it was terrific, that might well prove the deciding factor. But did he think it was terrific? I have no idea, 'cause there's no grade and I won't read his review for fear of learning more about the film than I'd prefer.

(2) I, too, used to think the 100-point scale ludicrous. And no, there's no significant qualitative difference between, say, a 76 and a 77—most of the time, the exact number I assign to a movie is fairly arbitrary (although some numbers are starting to develop distinct personalities, to my surprise—69, for example, has become my "film maudit" rating, applied to movies that don't ultimately work but have stretches/elements so exciting that I find myself wanting to pretend that they do). But it's silly to claim that additional gradations don't clarify matters. I'll show you what I mean.

Chasing Amy
Spun

Two films I've seen. Liked some parts of each, didn't like others. Let's evaluate them using a few different scales. First, the Maltin four-star scale:

Chasing Amy: **1/2
Spun: **1/2

So I liked them roughly the same, right? But now let's switch to letter grades:

Chasing Amy: B-
Spun: C+

Hmm. Clearly I liked Smith's film a little more. What happens when we switch to the 100-point scale?

Chasing Amy: 63
Spun: 46

Does it still seem like I value them equally?

md'a

Posted by: Mike D'Angelo at April 15, 2024 07:19 PM

Mike:

I think you lost track of what your own ratings mean. Let me explain. Here's the
mathematical mapping between the scale of 0-4 and the scale of 0-100:

0-4 Scale 0 - 100 Scale
-----------------------------
0 - 0.5 ....... 0 - 12.5
0.5 - 1 ....... 12.5 - 25.0
1 - 1.5 ....... 25.0 - 37.5
1.5 - 2 ....... 37.5 - 50.0
2 - 2.5 ....... 50.0 - 62.5
2.5 - 3 ....... 62.5 - 75.0
3 - 3.5 ....... 75.0 - 87.5
3.5 - 4 ....... 87.5 - 100.0

I'm assuming the mapping between the 0-4 scale and the 0-100 scale is linear.
If it ain't, then the use of the 0-100 scale is even more confusing and less
predictable than I originally thought.

Clearly, a rating of 2.5 is the upper bound of the 50 - 62.5 interval and
the bottom bound of the 62.5 - 75.0 interval. This would put a rating of 2.5
somewhere between 56.25 and 68.75. While your rating for "Chasing Amy" would
properly fall in this interval, your rating for "Spun" is *way* outside this
range. In fact, a rating of 46 falls in the 1.5 - 2 star rating which means
that you never should have given it a 2.5 in the first place. You would have
given it a 2-star rating, which (even on the 4-star scale) shows that you
clearly prefer "Chasing Amy" to it.

So "Spun" would have to be somewhere in 56.25 and 68.75 on the 100-point scale,
not a lowly 46. Even the lowest value in that range gives it a 56.3 which still
puts it close to the 62 of "Chasing Amy" which proves the point that more
resolution is not better.

Sorry to over-analyze the math, but that's what happens when you start using
such an over-resolved scale -- you yourself lose track of the rating mappings!

:)

Dan

Posted by: Dan at April 16, 2024 08:25 AM

Yeah, I noted the mathematical disparity too. But I'm certain that Mike knows full well that the numbers don't match up, and is trying to map qualitative, as well as quantitative, differences between various grading schemes. Seems to be saying that "** 1/2" is a broad category of opinion, sitting as it does between the implied mediocrity of a "**" movie and the implied value of a "***" movie -- it becomes a catch all for not-bad movies without enough merit to warrant that third full star.

I take part in a survey that Mike organizes at the end of every year that requires that I assign a star rating to every movie I've seen all year. This doesn't bother me. But I do not directly map my letter grades to star grades. In fact, I pay no attention at all to my original grades when I hand out the stars, which I'm sure results in some telling disparities between my opinion of a film immediately after viewing and my opinion of it at the end of the year. The thing is, grades don't map directly to stars. Some of my "B+" films would get 3.5 stars, others would get 3. Some of my "A-" films would get 3.5 stars, others would get 4.

The way I think about my opinions of movies changes, somewhat, depending on how I'm grading them. On a letter-grade scale, a "B" is a pretty solid grade. Movies I really liked get bumped to a "B+," which is my magic grade for movies that I really dig, and movies that I harbor reservations about get docked to a B-. That doesn't map so easily to star grades, where a "***" is a solid rating, but docking a film from that to "** 1/2" seems more harsh somehow. And while I guess "*** 1/2" implies the same near excellence that a "B+" signifies, there's lots of "A-" films that I wouldn't want to give the full four stars to, either.

And I don't know what the hell I'd do with a 100-point scale, but I appreciate Mike's further explanation of what he thinks he's doing with it.

Posted by: Bryant at April 16, 2024 10:00 AM

I think you hit the nail on the head. Rating systems force one to map their qualitative opinions into quantitative ones. No wonder we're all so confused!

Maybe you're right! Maybe the ratings have to go. Maybe we can still have a consumer guide by using qualitative words instead of mathematical ratings?

Excellent, very good, good, OK, avoid.

But does this put us back to square one again? I'm too confused to continue!

Posted by: Dan at April 16, 2024 12:36 PM

HI, Wow.. this is a very informative website!
I enjoy your site very much! Keep up the good work!

Posted by: South Beath Diet at February 1, 2024 02:37 PM

Post a comment




Remember Me?

(you may use HTML tags for style)