What changes racer motivation and activity?
Well, in my latest statistic-fever, after creating the "Sent Replays Analysis" I decided to supply you with some more detailed statistics derived from those information you can also find on the webpage.
Well, we always knew the number of racers on each track. It's enough to look at the scoreboard. Sure, some might think there are fake racers, but we just don't care about these people now. So lets assume we know the racer number.
On the other hand I knew always the number of replays sent on each track. But each replay sent is only recorded in the database since the 2004 season. That's why you can acccess the replay stats only from this point.
But I thought two finished seasons are good enough for a comparison, so I made it for you. How does racer number affect the racer activity? (e.g. the replays sent by racers.)
Here you are, a graph showing the racer number and the number of replays sent from C34 to C58:
What can we see? The received replay count started at a reasonably high value in 2004, but then it dropped and dropped and dropped though the whole year. It did this despite the increasing number of racers. Then it started to rise after it bottomed out on ZCT43, and made an exceptionally big jump up on ZCT45. This could be because it was a great track, or just racers were bored, I don't know. :)
At the end of 2014 and the beginning of 2005 as more racers came, more replays were sent. The all-time-high replay number was on ZCT49, nearly the double of thw usual number! Then strangely as the racer number was on the run in the middle of 2005, the replay count was going down again - but still - it was higher than the average in 2004.
Well all this is intresting, but show no relevant information to us, we don't know yet the racer activty. We can get this by dividing the number of replays with the number of racers. Easy, isn't it? We get the average number of replays per racer each month. Now let's compare this with the racer number again!
Well this is, a very interesting graph. The first thing we can notice, that there is no sign of a constant raise or fall. Usually the racer activity is around 3.5-5 replays per track.
Another thing we can see, that there is no strong connection between the racer number and their activity. I've also calculated a statistical value called coefficent of determination which was in this case 0.04. This means, that the racer number defines the racer activity only in 4%! Who would have thought? Lots of competitors don't make you more motivated to race.
I wanted to show you that this coefficent really works, and isn'T just a play with numbers (of course you know this if you learned statistics :)) so I have calculated two other values The coefficent of determination between the racer number and the number of received replays is 21%. Sure it's obvious that more racers send more replays, isn't it? :) And also the coefficent between racer activity and the number of replays is 52% This means that the number of replays landing in my inbox is determined 21% by the racer number, 52% by your motivation/activity, and 27% by other things we can't define.
Well, what is important then? Maybe a good track? Well I think a good track is always a subjective thing. Some like it, some dislike. Probably on good tracks more racers join, but it doesn't modify their activity too much. But of course it counts - you send more replays when you like the track. Still - we can't measure how much does a track affect your motivation. We could, if racers would rate tracks on a scale each month. Maybe we'll do that later. :) But what else can define the activity then?
Maybe racers are more active at some parts of the year? Is this possible? Remember, 2005 started with ZCT47, 2004 started with ZCT34. What can we see? There was a constant rise in the activity each year until march, the third track. Intresting. What's even more intresting, that both years this thir track was the absolute record in activity throughout the whole year. And then, the activity slowly falls down and down. Probably first because the summer exams and holidays are coming (will this change with the rising number of racers from the southern-hemisphere?)
Then suprisingly as the end of the season comes by, activity still remains at the usual level, only a few tracks are exceptional (like ZCT45 and ZCT56), but they don't affect the general level of activity, it quickly falls back to the normal level. You can also notice, that usually after these excetionally popular tracks the next track has a weaker activtiy than usual. Probably racers lose their motivation after having an usual race after a great one.
So what is the conclusion? Honestly, I don't really know. There is a significant sign, that the racer activity is thee highest in march, and still higher than usual through the whole spring. And we can conclude that racer number doesnt affect the activity, nor does it change as time passes by (in 2004 the average activty was 4.94, in 2005 it was 5.02).
So -most likely- everyone know how much time he wants to spend with playing Stunts. He knows that, and keeps that, and nothing can change this.
Well I'm fine with this. Until you keep sending your 5 replays per track in average, I'll be happy. And the race will be intresting. :)
Zak McKracken, Budapest 21.01.2006