It took me around 10 months to read 66 novels, produced over a little more than 55 years. It's been a lot of fun most of the time. While there were definitely some disappointments, I've generally looked forward to reading each novel.
Christie's level of mystery writing is generally very high, and now when I'm going to take a look at the rankings I've given to the novels, I want to emphasize that these novels are ranked against each other, not against any other mystery writing. If they were, there would have been too many novels with a ranking above 90.
Over on Ah Sweet Mystery, Brad's been doing a series of posts on Agatha Christie's production during different decades. His runthrough is excellent, and you should all take a look at it, if you haven't already. (The link above will take you to the first post in the series.)
This overview will be different (because there's no point in
competing with the internet's resident Christieologist), and I'll keep
this post focused on the ratings I gave to the novels, and what they
might tell us about Christie's career.
I'd like to start with a quick look at the ranking of all the novels. We'll take a closer, more detailed look further on, so this graph is just to get a quick overview over how my ratings look.
While I was doing my ratings, I never looked back on previous novels to see what I'd given them. I didn't want any influence from that. Now, when I see this graph, my first thought is that the overall trend is exactly what I expected, but I also think that I might have been a bit kinder to the late career works than to some of the early ones.
What's perhaps most noticeable is that during the middle of Christie's career, she had very few dips below 50. And from the 1960s on, she rarely got more than 60.
Highest rating: The Murder of Roger Ackroyd 93
Lowest rating: The Big Four 12
Average rating: 46.8
Median rating: 39
Well, let's take a closer look at Christie's first decade. I could call it a bit up and down, but that's hardly the whole truth. Looking at the ratings, there's one really high point, one very low, and the rest of the novels, while varying in their ratings, are generally somewhere in the middle, though mostly below average as can be seen by the median and average ratings. The decade is also flanked by two novels that had better ratings than the rest of the middling ones, The Mysterious Affair at Styles and The Seven Dials Mystery, which makes the others look a bit worse than they actually were.
The data here is what tells me that my ratings haven't been entirely consistent over this re-read, because I think several of the 60s works should be rated worse than quite a lot of these ones.
The ratings aren't helped either by the fact that there are a lot of thrillers here. In fact, I'd contend that there are just four true fair play mysteries among these ten novels, though I wouldn't argue with you if you pointed out that at least one of the adventure thrillers had quite a lot of fair play elements in it: Seven Dials, if you hadn't already guessed.
And since fair play mysteries were Christie's forte, it necessarily follows that the ratings of the works of this decade will suffer.
What the data also tells us - even if you don't agree with my individual ratings - is that this was a decade of learning, of stops and starts, and considering the personal upheaval that Christie went through immediately after publishing Ackroyd, the ratings following that one are not all that surprising.
It should be added here that a majority of Christie's short stories were produced in this decade, which shows that she was full of ideas already at this early stage.
Highest rating: And Then There Were None 99
Lowest rating: Why Didn't They Ask Evans? 44
Average rating: 74.8
Median rating: 78
Things look different here, as you can see. The lowest score here is just a smidgeon lower than the average of the 20s. While there are still valleys and ridges in this curve, they're generally more flattened out. The fact that one of the strongest novels, Murder on the Orient Express, is followed by the lowest rated makes for a sharp contrast, as does the fact that the highest rated one was preceded by one I had some reservations about, Murder Is Easy.
But just take a look at those scores. The average is at almost 75 out of 100, and the median is even higher, which tells us that there are a few titles that drag the average down. Add the fact that the 30s is the decade where Christie published the most mystery novels, and that's an awesome run.
Even more amazing is the fact that while the 20s featured the bulk of
her short story writing, this decade provided
most of the rest of them. (Yes, she wrote very few short stories in the decades
after this one.) While most aren't up to the standard of the novels,
it's amazing to think how prolific Christie was here.
The first novel of the decade gets it off to a running start by getting a higher rating than everything from the 20s (except Ackroyd), and then she just keeps 'em coming. Look at the trifecta of Nile, Appointment and Christmas - I wonder if there is any mystery writer who had a run like that. I'd have to think a while before finding one, at least.
Obviously, the ratings are helped by the fact that Christie stayed almost completely away from thriller writing. Evans is her only dip into that sub-genre, and well, its rating says it all. I think she had realised that her forte was her mystery writing, not her thrillers.
If the 1920s were a decade of learning and getting over personal tribulations, this decade shows that Christie had not only learned but mastered her craft, and the novels we find here will wow new readers for quite some time yet.
Highest rating: Evil Under the Sun 95
Lowest rating: N or M? 47
Average rating: 73.8
Median rating: 75
The variance is lower here, but otherwise the numbers are still wonderfully high. There's a rather big dip with N or M? and The Body in the Library (which looks even worse because of the two novels surrounding that period), though in fact, the actual ratings are just a tad below 50, but otherwise the ratings are quite high. Look at the end of the decade as well, that curve looks very promising.
I mentioned during the re-read that in the 1940s, Christie changed her style somewhat. (In fact, I'd argue that it began with her last title of the 30s.) Many of her plots began to rest on her characters and their traits. Don't get me wrong - we're not talking about modern "psychological mysteries" here. We're still generally firmly rooted in fair play writing. Clues abound for the reader to find (or not), the difference is simply what type of clue Christie provides us with.
Christie again managed to stay almost entirely away from thriller writing, with one exception, again the lowest rated novel of the decade...
What I take away from this decade is that Christie had mastered her craft during the 30s, and here she branched out a little, offering stronger characterisation with her plots while generally keeping the standard as high as before.
Highest rating: A Pocketful of Rye 89
Lowest rating: Cat Among the Pigeons 44
Average rating: 65.6
Median rating: 62
That might be the most telling graph here...
While the 1950s started out well and the highest rated book of the decade was actually published four years in, the trend over the decade is clear. It also looks as though the first half of the decade is generally much better than the second half. But it's also worth remembering that the lowest rated novel here actually is on a par with the lowest ones from both the 30s and 40s, so not even the numbers for the last half of the decade are that bad.
So, while the trend is noticeable - Christie's powers are starting to wane - my ratings suggest that even if you pick a novel at random from this decade, you will still find a worthwhile read. To be even more specific, I'd argue that if you pick a book from the early 50s, you won't be
disappointed, but if you choose something from the second half of the
decade, things get a little bit more iffy. Something happened around the writing of Hickory Dickory Dock, and after that there wasn't much gas left in the tank for Christie.
It is, however, also worth keeping in mind that Christie, probably to spice things up again and not get stuck in a rote, returned to her old love, adventure thrillers. She was never particularly great at writing them, but still put one out each decade, and here we got two of them. Still, it's only fair to point out that these two are the best of the bunch.
The 1960s and 1970s
Highest rating: The Mirror Crack'd from Side to Side 72
Lowest rating: Passenger to Frankfurt 2
Average rating: 39.8
Median rating: 37
It's interesting to note that once again, a Miss Marple novel is the highest rated novel of this period.
But yeah, that's overall quite a dip. Those average and median ratings don't look good. Though the 1920s featured quite similar ratings, I think my ratings were kinder here, as I said above.
Remember, though, that the average is also quite heavily affected by my personal disdain for Endless Night, a novel many others have very warm words for. And obviously it doesn't help that Christie's two worst works by far were published during these last 15 years or so.
Looking at the graph, it almost seems that you'd be fairly safe if you chose every other book here. The graph goes up and down and up and down with some regularity...
While the final two publications give the curve a healthier look, neither of them are rated better than third in this time period. It's just that the run from Endless Night to Postern of Fate is so bad, they seem like a return to form. If they'd actually appeared around the time they were written, they'd have been distinctly below average. But in the early 60s, when Christie wanted to, she could still put out a solid mystery.
Some random statistics
There are six novels with a ranking above 90:
The Murder of Roger Ackroyd, Murder on the Orient Express, Death on the Nile, And Then There Were None, Evil Under the Sun, Crooked House
There are likewise six novels ranked lower than 30:
Passenger to Frankfurt, Postern of Fate, Endless Night, The Big Four, At Bertram's Hotel, By the Pricking of My Thumbs
I suppose there's some symmetry there, though I had to raise the bar to 30 for the lower ranking ones. There are four Poirots among the top rated novels, and two standalones. For the lower rated novels, there are two featuring Tommy & Tuppence, one each for Poirot and Marple, and two standalones.
Which leads me to this final bit, where I look at the average ratings for the characters' novels.
Hercule Poirot average rating: 68.0
Miss Marple average rating: 63.25
Tommy & Tuppence average rating: 28.5
Inspector Battle average rating: 65
Colonel Race average rating: 66
Standalone novels rating: 53.5
Poor old T&T, Christie didn't help them by featuring them twice in her late career.
Battle and Race are both helped by having been in Cards on the Table (and Race even more by having featured in Death on the Nile).
The average rating for Poirot is absolutely fantastic, seeing as he has 33 novels, but Miss Marple's average isn't bad either.
While I'm certain that none of you reading this will agree completely with me on the individual ratings, I hope (and think) that you concur with my overall findings.
Christie's 20s were a decade of learning that showed us that she had oodles of ideas, even if the execution wasn't always up to scratch.
By the 30s, she knew exactly what she was doing, she was firing on all cylinders and produced a volume of work that will stand on the top shelf of mystery writing for eternity.
She switched gears coming into the 40s, providing us with greater characterisation without skimping the least on her magical plotting, giving us another set of excellent mysteries.
As the 50s came along, Christie was still at the top of her game, but by the second half of the decade she was flagging a little. Individual titles could still be very worthwhile, but the greatest highs simply weren't there any more.
The 60s and worse yet, the 70s are better approached with care. While many of the titles from this period still have interesting ideas or characters, or an exciting turn of events, there are very few of them where the entire execution is truly worthwhile. And unfortunately, there is the occasional title that is better left on the shelf and not read for any other reason than completism.
That's it for this statistics feast. There's another post in the offing, where I'll discuss one or two thoughts or theories that I've come up with during the re-read. Don't miss it!