I agree with the Kruel's assessment of scoring.
I also think both sides are saying the same thing, they just don't realize it because of differences in familiarity and approach.
What the current scoring system lacks is context. The current ranks are determined solely by numeric averages and nothing more.
Let's look at an example: Someone can come in from the top of a massive amount of inbound enemies, kill 10 planes in under 10 minutes, and then RTB. Or, even better, let's say 10 bombers. Adding in an additional 10 minutes of travel each way for ingress and egress, that brings the total sortie time to 30 minutes. Now let's look at the stats.
Total flight time: 30 minutes
Total Kills: 10
Total Deaths: 0
Total k/d: 10.0
Total Sorties: 1
Kills Hit Percentage: (let's figure 17.5%, for killing bombers - not unreasonable by any means)
Kills/Time: 20/hour.
Kills/sortie: 10
Kill Points: ?
So using the completely realistic scenario above, we end up with #1-worthy stats in all categories, except kill points. This person would undoubtedly finish top 20, probably top 10, with a very good chance of being top 5.
All from one sortie.
The above stats are impressive, when taken out of context. Within context, they're less than impressive - someone had a really good, single run killing bombers. The problem, then, is that the current system measures statistics, not performance.
I see many people here arguing that statistics in and of themselves do not give an indicator of skill. I completely agree, which is why I think we should change the system - hear me out.
Let's view another pilot with these stats:
Total flight time: 5 hours
Total Kills: 50
Total Deaths: 10
Total k/d: 5.0
Total Sorties: 10
Kills Hit Percentage: 9%
Kills/Time: 10/hour.
Kills/sortie: 5
This guy will probably finish top 50, maybe top 25. However, these stats, individually, are only decent at best, so his overall ranking will depend on the law of averages - nothing more.
But the real question here, is how does this guy, which we will refer to as pilot B, compare against the pilot above, pilot A? The answer: We don't know, and can't tell, from the stats.
The current stats are worthless for anything except numbers right now. True, you can infer a certain amount of skill for those that consistently rank high, but outside of inferring skill based on long-term consistency, they're worthless in and of themselves.
Now, let's go back to pilot A. He had a really good sortie. A really, really good sortie that he intends to repeat. Let's pretend that he rolls again... goes on a 30-minute sortie... and is shot down by an average-skilled player. Let's assume he shot several times at his opponent throughout the fight, lowering his hit percentage to 9%. He now ranks lower than pilot B, as all stats are identical, with the exception of kill points, which pilot B should have a lead in.
However, we're just pretending. Pilot A didn't roll again, didn't get shot down, and now he's parked his name at the top of the leaderboard as it's actually a shade account that he rarely flies. One of the top 25 spots, and most likely top 10 spots, is now taken due to an extremely lucky run. Now, we know that he's only marginally skilled, but the stat system doesn't weight performance, so he will remain ranked over pilot B who clearly has a much more solid gameplay history and skill set.
The problem that I see with the system isn't the rankings or the averages, it's the lack of a weighting system that gives some type of context. Now, we have to be careful here, because we don't want subjective context, we want objective context. We don't want to try to calculate if it was a 5v1 fight vs a 1v5 fight, or if player A got kills from HO's or not - it's ridiculous to try to adjust for those factors.
What there is improvement for is quantity of quality. For example, someone with a 10:1 with 50 kills should be given some sort of weight against someone with a 10:1 with 40 kills. However, due to score adjusting for imbalanced sides, it's realistically possible for the 40-kill player to gain more kill points, especially based on the aircraft that they fly. In fact, for example, someone furballing in early/mid-war Spitfires can use this to game the system.
People claim that the Kill Points stat balances out the above, but that's not the case. Kill points acts independently, as its own stat, so instead of directly affecting the weighting of kills/deaths, it indirectly affects it through overall ranking. It doesn't directly modify the ranking of K/D or Kills per Sortie, so it doesn't really work as a weight of quality.
The same thing could be said of accuracy. There should be some type of weighting for the guy that fires 150,000 rounds at 9.9% accuracy compared to the guy who fired 1000 rounds at 10% accuracy. The current system negates long-term, consistent performance and instead rewards short term luck.
As it stands, the scoring system is all about gaming numbers, not improving performance. By weighting kills, accuracy, etc., it downplays the advantage of being able to run one or two or five sorties and then retiring for the month, as those sorties would have relatively little weight. An adjusted scoring system would reward long-term, consistent, good performance over someone who had one "lucky" great run.
What the changes would not do is change the standing of good or even great pilots. Pilots who are consistently good could fly, no longer fearing a freak-chance death from a .50-cal pilot wound at 1200 yards. Deaths and misses are not primary concerns, because they are weighted out in the long run. It would encourage people to engage and fight more, instead of running the second it looks like the tables might turn and the fight could end up with a single death that would wreck that player's monthly stats.
Poor pilots would remain being ranked where they are currently at. There are no suggested changes that would cause them to rank higher.
Fighter Ace had similar mechanisms, where stats could be pulled based on volume of kills, deaths, sorties, etc. What resulted from this is that no one, literally no one, cared about the players' stats with under X amount of kills. The community that resulted from it was one that was focused on long-term performance statistics. No one cared if you had a great week or even month, they cared how you performed over the long-haul. "Who are the best pilots with more than 10,000 kills?" "Who has the highest kills per hour, with more than 1,000 overall kills?" - those were the types of scoring inquiries that resulted from the changes, and they were all positive.
I believe these changes would bring about positive change, as players would be focused on long-term stats and on finding fights, as volume and consistency matter more than single, lucky, low-volume runs.
If you made it this far, thanks for reading.
