Pigeon Ascent Analytics Analysis
Hi there! I’ve been away from game development for a while, and up until a few days ago, I actually forgot we were collecting Pigeon Ascent analytics. I’m not a data scientist or anything like that, but I managed to make some plots, and in this post I’ll be listing my conclusions from them.
It’s possible I’ll have some wrong conclusions, but that’s part of learning, I guess. If you notice things I didn’t, feel free to send us an e-mail, or drop a comment on the itch.io game page, whatever you prefer.
Amount of runs won/lost
So, the amount of wins and losses seem somewhat balanced, with a slight tendency to wins, which I think is something good. If the data was too heavily skewed torwards losses, the game could be frustrating, and torwards wins, the game would be too easy and boring.
An argument could be made that certain games are fun and have a high loss rating: games like Elden Ring come to mind. Based on Steam achievements, at least about 28% of all players managed to finish the game with a certain ending, and Elden Ring is a pretty good game, of course. However, Elden Ring also heavily relies on the player’s skill, to a point that no hit runs and level one runs are possible. Pigeon Ascent, on the other hand, is an autofighter game, and some luck inevitably plays some role in how the game turns out. If the losses percentage was higher, the game would be immensily frustrating I think.
Also, I know it is bizarre to compare Elden Ring and Pigeon Ascent. They are on completely different leagues. But I do think the point still stands.
Now that I think of it, I wonder if having the odds close to 50/50 means that luck is playing too big of a role… I didn’t consider this yet… My immediate answer would be that yes, but at the same time, following a consistent “build” can lead to wins more easily, and choosing your opponents do help as well… therefore, there must be at least some element of skill in how the game progresses. Maybe this is why victories have a slight upper hand compared to losses.
Evolutions that won/lost the most
I should’ve displayed a percentage in those plots… but too late now. Anyway, done is better than perfect (but I do realize this saying doesn’t really make sense).
Well, the first thing that I notice is that the Normal pigeon has a ton of losses. This is likely explained by all runs starting with it, so by far it’s the most common pigeon type. Since it’s the most common, therefore, it might end up having the most losses too. Still, this could suggest that the early game is too rough…
Another thing I notice is that the strength pigeons are really up there in both plots. This could also be explained by they being the most common evolution. At least I think so.
Besides those two things…
- The platypus pigeon is way too strong. Likely due to its passive skill, which raises its stats a lot. Still, it’s likely to be the most common pigeon, based on playtests I saw. This happens because players usually try to keep their pigeons balanced.
- Speed focused pigeons are the ones suffering the most to win the game.
- Defense pigeons are seemingly doing OK, but the Fridgeon evolution seems quite weak compared to the crusader.
HP values that won/lost the most
For wins, there seems to be a good spread… the shape do reminds me of a bell curve, with a mean value of around 70. Not sure why there’s a dip on 80 though. Maybe by then some players prefer investing on something else. I should’ve created a FOOD plot as well.
For losses, interestingly enough there’s also a kind of bell curve IMO, but there’s also a good amoung of points in the 5-9 bucket. This is likely because of losses in the early game… maybe players tried a build without a ton of HP investment. Or maybe, as I commented earlier, the early game is too rough. Maybe people are just learning to play the game by then. Having a dip right after this bucket suggests this could be the case.
Session time
This was very interesting. The average playtime is around 6 minutes… which makes sense, the game is pretty short after all. There are speedruns that clock below 1 minute, in fact. However, there’s lots of people that play for way longer… a +15 minute session isn’t that rare, which is very interesting to me. I say this because those analytics were a way for me and Sucraiso to gauge where the game is rough, and what’s good so far. Now is a good time to go to my conclusions, I think.
Conclusions
We’ve been thinking of making Pigeon Ascent 2 – or at least a better version of the current Pigeon Ascent, one that we would be happy enough with the quality to the point of asking for a price – for a good while now. But until then, the plan was to polish the Pigeon Ascent we have now so that it becomes essentialy the demo for the new version.
But based on what we collected so far, we are not sure on where to improve things. Some more balancing would be better, yes, but the way I see it, the game would change too much. I was planning on adding skills, equipments, and other things, but the game could end up being too different. The way it is right now, it seems to be in a very delicate balance, and I’m not sure I can improve it much more without starting to remake things.
For instance, having strength, defense, and speed as stats would likely end. We would have to add something else… which could change a good chunk of how the game works. Doing this is not out of question, I’m just unsure so far how to tackle it. It could end up being left entirely for the sequel. I’m not sure yet.
Pigeon Ascent should receive some new updates this year: not sure if new features will come in, but I should at least fix some bugs and improve some other things. We’ll see… I hope you keep enjoying to play Pigeon Ascent! Any feedback is very welcome, as always.