Recently, a friend of mine sent an email to several of us saying “Sentinels of the Multiverse is on sale for Android”. Cool. I was going to forward to some other friends, but then realized … they don’t like Sentinels. And in fact, all of my friends who don’t like Sentinels seem to like Seven Wonders.
In fact, it seems, in my game groups, Sentinels and Seven Wonders are anti-correlated: If you like Sentinels, you don’t like Seven Wonders. If you like Seven Wonders, you don’t like Sentinels.
Is this true in general? Or is it just a blip in our game group?
If we could just tap into BoardGameGeek, find everyone that rated both games and compare them, maybe we could see if this were a real trend! But, alas, we can’t do that. Or can we?
My friend Josh pointed out that BoardGameGeek allows an interface to download just such data:
With this interface, Josh downloaded what we needed (he used a curl script to gobble the data and parsed the data with a Python script) and generated some stats and graphs. It was easy: he did it in about an hour.
So what did we find? In general, people who like Sentinels tend to like Seven Wonders, and people who like Seven Wonders tend to like Sentinels. There DID NOT seem to be the anti-correlation that we thought! Apparently, my little game group is its own little microcosm.
Here’s the data graphed:
On the left axis are the ratings (from 1-10) for Sentinels, on the right axis are the ratings (from 1-10) for Seven Wonders. The height of the 3D graph shows how many BGG users rated the game (on BGG’s scale of 1 to 10, where 10 is the best). You can see that the data seems pretty consistently rising on both sides. People who like Sentinels tend to like Seven Wonders and vice-versa.
Other interesting data from the time of the comparison:
- Seven Wonders had 61037 reviews
- Sentinels of the Multiverse had 12425 reviews
- There were 7880 overlapping reviews (i.e., users who reviewed both)
I am curious, what do other people see?