Forum
TF2 Player Rankings website - e-peens at the ready
Created 10th January 2015 @ 23:38
Add A Reply Pages: « Previous 1 ... 3 4 5 ... 14 Next »
Quoted from MARS^
How often does this update?
And can you implement something so you can see how much you improved/got worse, e.g. rank last week->rank this week?
Updating is a bit laggy due to caching, it can be up to an hour or more behind but if something goes wrong with the updating function then there could be a big gap until I notice. I’m looking at improving this.
I’ve got rating over time graphs on the roadmap.
Quoted from Bloodyyy
My k/d is on every class over 2,5 – 3,0. Yet im still with most classes #1000 ranked xd. I even have better stats as medic, then the MAIN medic of #5.
This should give you a clue as to the importance of KD. Anyone trying to improve their score will fail if they focus on a simple metric like that, unless they also happen to hit on the system’s idea of perfect gameplay by chance.
People got so obsessed by stats, so we got a ranking system that doesn’t make any sense. What’s the point of this? Why can’t we just gather logs and filter it by ka/d, dpm etc. That would say a lot more than these ranks.
meh. TF2 can`t have a working ranking system. There goes way too much into it that logs can`t show.
Quoted from Spreijer
People got so obsessed by stats, so we got a ranking system that doesn’t make any sense. What’s the point of this? Why can’t we just gather logs and filter it by ka/d, dpm etc. That would say a lot more than these ranks.
Bear in mind that these ratings are for lobbies and pickups, in which the best players often don’t play to win. These ratings tell you which team is more likely to win a lobby/pickup.
Quoted from Spreijer
People got so obsessed by stats, so we got a ranking system that doesn’t make any sense. What’s the point of this? Why can’t we just gather logs and filter it by ka/d, dpm etc. That would say a lot more than these ranks.
The original project was designed to provide a method for lobby/pickup sites to run their own team balancing. Despite handing over copious findings the limitations of developer availability in those services meant that they didn’t get developed very far if at all, so I took it upon myself to produce something that would minimise the workload for them.
Because there is secrecy necessary in this process, and because stats are relevant to the strength of your opponents, there isn’t a clear link between rank and stats displayed with each player. This is partially a design issue, I need to consider how to weaken the implicit link.
Having compiled this massive database it occurred to me that it would be interesting for players to see it, as it would be impractical for me to also continuously pass over masses of ranked data to services to present themselves.
Zoob is also planning a big update to logs.tf that will provide a lot of features for compiling and filtering your own collections of logs, so it made no sense at all for me to try to clone that effort.
So that’s the point of it, and why some elements may not appear to make much sense superficially. I would say individual stats can say some things, but if you only look for high KD and DPM then you will be penalising players who look to challenge themselves by taking on the best. This system doesn’t do that, but with anything that complex presenting things clearly is a big challenge whilst maintaining the confidentiality of the ranking method at the same time.
is this the system used to scramble teams on pickup?
also please add this to the tf2center page, similar to the stars script kengur made
Quoted from letto
is this the system used to scramble teams on pickup?
also please add this to the tf2center page, similar to the stars script kengur made
I’m not sure what’s going on with the TF2 Pickup balancer, I know there was an early version used at one point but I’m not sure what’s going on now. If they want to/can try using this one then they’re more than welcome.
Quoted from Gentleman Jon
If they want to/can try using this one then they’re more than welcome.
They should, their balancing system really is the worst
Quoted from Gentleman Jon
[…]Because there is secrecy necessary in this process[…]
There isn’t. Why would there be? Only reason would be bad design. Enough ranking systems exist that are open source or well described.
Quoted from InstantMuffin
There isn’t. Why would there be? Only reason would be bad design. Enough ranking systems exist that are open source or well described.
The reason is because people will try to game the system, which is why Valve doesn’t release details of their ranking systems. For now I doubt people care enough to bother, but if the ranks started being used for something concrete such as a gateway to higher level games attempts to gain high ranks without earning them would begin in earnest.
Open-source ranking would more easily be abused, especially in TF2. People would just be fixated on that one task that gives them e-pen points (for example, meds would be so afraid to drop they’d pop in the most absurd locations if they knew how many points they lost)
Quoted from Burn
Open-source ranking would more easily be abused, especially in TF2. People would just be fixated on that one task that gives them e-pen points (for example, meds would be so afraid to drop they’d pop in the most absurd locations if they knew how many points they lost)
That’s like saying open source encryption is bad. Either it’s bad by design or it’s bad by implementation.But it is bad. In a way open-source (and by an extent free software) is even required tor a proof of reliability (I’m actually quoting Richard Stallman on this). Only thing the valve point does is showing that valve’s csgo ranking system is bad by their own estimation.
Now I fully understand aspects being trade secrets to prevent rip-offs, but saying reliability is fully dependent on the secrecy is like saying good encryption relies on secrecy of the encryption process and not secrecy of the key. It’s bs every computer science freshman knows. Just be honest then, no one is different in that regard. Work deserves recognition and in a way sometimes keeping stuff to yourself is the only way to get proper credit.
Open sourcing relies on the fact that anyone can adjust or report the invalid codebase. At a ranking system it’s primarily used to game the system.
In this case, security through obscurity works.
He can’t adjust his code? Anyone can do adjustments by doing their own implementation. It’s not about securing the implementation itself, it’s about making the system solid by making it a publicly proven concept. Which would be a necessity to prove anything as solid.
A solid ranking system cannot be gamed. Therefore its specifics can be open. If exploits will be found, the system can be adjusted. Which is even more wise to do in early stages to prevent loss of data by time spent collecting and evaluating false data in an evidently false way. Because if leaks are found, who says the data was valid in the first place and stuff hasn’t been abused? First principle of a compromised system, it can’t be trusted.
Even wikipedia knows this: http://en.wikipedia.org/wiki/Security_through_obscurity
Last edited by InstantMuffin,
Add A Reply Pages: « Previous 1 ... 3 4 5 ... 14 Next »
content rss