How I Rate Movies

IMDb has a 10-star movie rating system. This allows users to vote with a whole range of different appreciations. However, as far as I know, there is nowhere any explanation of what a score of ‘3’ or ‘9’ is supposed to mean. IMDb leaves it up to the user to decide how to rate a movie. Many people seem only able to give ratings of either 1 or 10, which is why you'll often see peaks in the vote histogram there. Most other people are still biased towards the lowest and especially highest scores. I've seen reviews where people summed up a whole list of bad points about a movie, and still gave it a 10. Eh?

In an attempt to arrive at a fair and balanced rating system, I created the guidelines below. Actually the only reason why I put this online is so I always have access to this ‘cheat sheet’ when I want to rate a movie. It is not my goal to impose this rating system on anyone, but if you've always felt unsure about what score to give a movie, it could be a useful guideline. Just don't assume that everyone rates movies according to these rules.

1. ‘Do Not Want’
God awful, makes you want to gouge your eyeballs out with a spork, and either head-butt the TV or try to hit the cinema screen with projectile vomit. Everything about this movie is bad to such a degree that it doesn't even become good in its badness. You really wish you had done something more worthwhile during the movie's running time, like trimming your nose hairs. If someone would give you this movie, you would microwave, burn, blend or eat it, to avoid the risk that other human beings could be exposed to it.
2. ‘Awful’
Still awful, but has at least one thing that is done well, like one decent scare in an otherwise pathetic horror movie, one good laugh in an otherwise decidedly unfunny comedy, one clever plot element, etc. You would never ever want to watch this again except maybe for that single good part.
3. ‘Bad’
Bad, but you agree that watching this film was an OK pastime on a lazy weekend evening because there was nothing else on TV and you were too lazy to dig up anything better. Or, you felt the need to expose yourself to something crappy to recalibrate your appreciation for movies, and without being a totally shameful waste of time it reminded you how bad a film can be. You could have better spent your time, though. You will definitely avoid watching it again, even on the next lazy weekend evening.
4. ‘Nice Try, But No Cigar’
Still bad, but is ‘almost there’. Either it has some good parts that are ruined by bad parts, or it stays at a constant level of “it had promise but the good part never came”. You still would never want to watch it again.
5. ‘Meh’
The threshold for ‘OK’. It's not good, not bad, just acceptable. This is the kind of movie that only just makes you feel you didn't waste 90+ minutes and/or the price of a cinema ticket. You would only want to watch it again under the conditions of 3, but you would never ever consider doing more effort than pushing a button on your TV remote to watch it again. This movie is either an equal mix of good and bad parts, or is just so forgettable that each time someone mentions the title, you need to read the plot and look at screenshots to remember what it was about. If you would be given this movie as a present, you would sell it or give it away.
6. ‘Not Bad’
It has some aspects that lift it above mediocrity and make it quite enjoyable to watch, but it either never becomes really good, or if it does, it still has some bad parts that drag it down. You would only watch it again spontaneously if it were a long time ago since you saw it and you re-watched all your movies scoring 7+ too recently. You would only recommend this to someone if they're really into the genre, but you would still warn them that it's not that good. You would never buy it, but if someone would give it to you, you wouldn't bother selling it unless you need to make room or are in desperate need to gain a few bucks.
7. ‘Good’
This movie is really worth watching and you would watch it again spontaneously, but not too often. It's just not that good that you would say to your friends that they would really miss out on something if they didn't watch it. It's very enjoyable despite some noticeably uninteresting parts, subpar acting, plot holes or other negative points that you'll always remember when thinking about this film. If you would find it in the bargain bin of your media store for a bottom price, you might buy it.
8. ‘Very good’
This is the kind of movie that you could watch again several times, even though it is clearly not perfect. There are still some minuses about this film, but the rest is good enough to almost forget those. You would recommend it to friends unless you know it's not their cup of tea. You would be prepared to pay the normal price in the media store, or buy it immediately if it's discounted.
9. ‘Excellent’
It is almost perfect. You could watch this movie again almost an unlimited number of times. It still has something significant you don't like, therefore you don't give it an outright 10. You would recommend this movie to your friends even if you know they don't like the genre, maybe it could change their mind. You would be prepared to pay more than the average price in the media store.
10. ‘Perfect’
You could watch this movie until eternity without ever getting bored by it. With each viewing you discover something new or see something in a different light, and/or the key scenes still thrill you even though you know them by heart. This movie has nothing in it that is less than good. If there's anything that is not perfect, it is greatly compensated for by something else that is stunningly brilliant or gripping. You would recommend this movie to everyone, even total strangers. You would not be satisfied with a simple regular release when buying it, but you would seek out a special edition in a fancy box or wait for the ultimate director's cut and be willing to pay a premium price.

Since the lowest scores are probably the hardest to decide upon, here is some more clarification. Films that score 1 or 2 are the ones that would make you walk out of the cinema and demand a refund, or switch the channel on the TV (or in retrospect, make you wish you had done that). 3 is just not bad enough to make you walk out of the cinema, but you'll still feel ripped off, the same goes for 4 but to a lesser degree. When watching a 3 or 4 movie on TV, you would only not change the channel if you knew there was nothing better on. 5 is the threshold for not feeling ripped off.

Of course this scoring system is not strict. A movie that perfectly matches one of these descriptions could still get +1 added to its score if it has something that lifts it higher, like being the first movie to introduce a really original plot element, or bringing an important message in an unobtrusive way. Or, it could have points subtracted if it has something that really offends you even though the rest of the movie is really good.

You should however never completely pan a movie just because of a single negative point. You should also not increase or decrease your score “to balance out other reviewers”, that's not how a voting scheme is supposed to work. You should especially not subtract points because for instance a 1973 sci-fi film shows outdated predictions of the future. If you don't understand why, you should stick to watching recent films that fit within your small present-day universe.

Do not rely on the average rating alone

Due to lack of a clear definition of the ratings, probable manipulation of the rating system by film studios, and the skewed way in which the average consumer casts their votes, it is pointless to judge a film by its IMDb rating alone. The voting histogram accessible by clicking the user count underneath the stars, is a lot more helpful.

The first histogram shows how many votes were cast for each score. The largest peak and the median are generally better indicators than the average. The second histogram shows average votes for various population groups. Every person will have to learn to look for personal indications in this histogram that a movie might either be great or to avoid. For instance, in my case “Females under 18” and/or “Females Aged 45+” giving significantly higher scores than other groups, is often a sign to stay away.

Do not rely on the ‘useful’ rating of reviews

The whole problem with the ‘useful’ meta-rating for reviews, is that the system itself is not useful by any stretch. It is implemented in a naïve manner that encourages manipulation, and generally causes the first few reviews ever posted for a film, to appear as the most useful reviews forever. The system calculates a weight based on the number of useful versus not-useful votes, and uses this weight to favour the most ‘useful’ reviews to appear on the film's main page, and for other ranking purposes. The usefulness estimate for a review will plummet instantly if the first few voters thumb it down. Unless the system only has a small number of reviews to choose from because only few have been posted yet, the thumbed-down review will then have no chance to compete with other reviews to appear on the main page, and be stuck in a digital oubliette forever.

This means it is worth it for a studio to hire an army of drones who regularly skim through recently added reviews for their own precious productions, and vote up/down the positive/negative reviews. Of course IMDb prohibits this, but making this undetectable is peanuts, therefore I am certain it happens. You can try it yourself: do your utmost best to write a review for a recent blockbuster that perfectly explains why you liked or disliked it. Wait one week and then look at how many ‘useful’ votes your review received. If your review praised the film, chances are your review will have been voted near 100% useful. If it panned the film, your review will likely have been panned as well. If it is neither scathing nor full of praise, you might get something in between.

Of course, it is not unthinkable that the same drones are also encouraged to apply the inverse strategy to films of rivalling studios. Moreover, there are obviously also quite a few visitors who vote honestly. The situation will therefore never be as black-and-white as I picture it above. Still, if IMDb would ignore and hide the ‘useful’ score until it has some statistic significance, and give reviews a fair chance until they are more certain to be crap, the whole review system in itself would be a lot more helpful.

The bottom line is: don't even bother reading the review on a film's main page: laugh in the face of the film studio drones and ignore it. Immediately click through to “See all user reviews” and use the ‘filter’ function. I really like the ‘Love/Hate’ setting which pits the most positive against the most negative reviews. Also skim through the reviews with the filter set to ‘Chronological’: this will give you a better idea of how reviews are distributed.

©2010-2016 Alexander Thomas