196 replies to this topic

[Der Geisterbaer]

[redacted] How about the current leaderboard?




looks like that +1 occurs quite regularly and not just to myself. So didn't I actually play 177 games? Did I lose or win 1 more often?

Edited by McValium, 24 March 2018 - 01:16 PM.
Redacted nonconstructive

[Bombast]

[redacted]

Quote

And where did I insinuate anything like that? Truth is: I didn't. As for the screenshot? How about the current leaderboard?


looks like that +1 occurs quite regularly and not just to myself. So didn't I actually play 177 games? Did I lose or win 1 more often?

You do realize what a draw is, yes?

Edited by McValium, 24 March 2018 - 01:17 PM.
remove nonconstructive

[Der Geisterbaer]

[redacted]

Bombast, on 23 March 2018 - 12:04 PM, said:

You do realize what a draw is, yes?

So you'd claim that all of these deviations come from draws? Let's ignore the fact that I know for sure that I haven't had a draw during this season and thus my +1 doesn't represent one: How often do you see draws that would explain so many +1 on just a single (not so random) page of the leaderboard. I for one experience "draws" like once or twice a year ... but hey, my memory is faulty and that's just andecdotal in your book.

Edited by McValium, 24 March 2018 - 01:18 PM.
removed nonconstructive

[process]

Draws are vanishingly rare from what I can recall, but keep in mind every draw hits 24 players at a time.

[Bombast]

[redacted]

Quote

So you'd claim that all of these deviations come from draws? Let's ignore the fact that I know for sure that I haven't had a draw during this season and thus my +1 doesn't represent one: How often do you see draws that would explain so many +1 on just a single (not so random) page of the leaderboard. I for one experience "draws" like once or twice a year ... but hey, my memory is faulty and that's just andecdotal in your book.

That's ancedotal in everyone's book.

Personally, I get a draw a month on average. Sometimes more. That's 24 people a month that get a draw on their stats at least. Or at least it would be, if it wasn't ancedotal, of course.

And 177 is a lot of games, actually. More, games more chances for a draw. You're on a page with a higher 'risk' of drawing, so none of that is a surprise.

Edit: Now that I think about it, I'm not sure I've drawn this month. Been playing 99% Hellbringer though, so it's all kind of a blur.

Edit 2: Hm, seems I overestimated my draw rate. Stupid brain.

Edited by McValium, 24 March 2018 - 01:19 PM.
removed nonconstructive

[Zergling]

Disconnects don't count as either a win or a loss, but do count as a battle played.

Although for it to not count as a win/loss, I think the disconnection has to occur fairly early in the battle, so winrate can't be padded by deliberately disconnecting when a battle is being lost.

And I'm not sure if it counts as a win/loss if the player reconnects before the battle is over.

Edited by Zergling, 23 March 2018 - 12:20 PM.

[Der Geisterbaer]

process, on 23 March 2018 - 12:12 PM, said:

Draws are vanishingly rare from what I can recall, but keep in mind every draw hits 24 players at a time.

I'm well aware of that. Question remains: Does he really think that those deviations can be attributed to "draws" only? Poor MOBAjobg who suffered 9, BRADTOTHEBONE with 10 and pyromant with 16 in just this season. And my +1 is still not explained unless "someone" would like to claim that I'm lying just to "be right".

[Bombast]

Zergling, on 23 March 2018 - 12:17 PM, said:

Disconnects don't count as either a win or a loss, but do count as a battle played.

Although for it to not count as a win/loss, I think the disconnection has to occur fairly early in the battle, so winrate can't be padded by deliberately disconnecting when a battle is being lost.

And I'm not sure if it counts as a win/loss if the player reconnects before the battle is over.

Kind of weird draws don't count as a loss for WLR, is t it? I know WoT counts them that way, I think Warthunder does...

[MischiefSC]

Zergling, on 23 March 2018 - 12:17 PM, said:

Disconnects don't count as either a win or a loss, but do count as a battle played.

Although for it to not count as a win/loss, I think the disconnection has to occur fairly early in the battle, so winrate can't be padded by deliberately disconnecting when a battle is being lost.

And I'm not sure if it counts as a win/loss if the player reconnects before the battle is over.

Yeah. A draw or any match in which you were disconnected for too long (is it 3 minutes? I forget. It's not much) counts as a match but doesn't reflect in your stats.

[Der Geisterbaer]

[redacted]

Bombast, on 23 March 2018 - 12:16 PM, said:

That's ancedotal in everyone's book.

Which part exactly? The one were I haven't had a draw this season (but certainly one the season prior)?
And if it's not a draw then it's a disconnect, right? Damn .. didn't have one of those either this month

Edited by McValium, 24 March 2018 - 01:20 PM.
removed nonconstructive

[Bombast]

Der Geisterbaer, on 23 March 2018 - 12:23 PM, said:

I'm well aware of that. Question remains: Does he really think that those deviations can be attributed to "draws" only? Poor MOBAjobg who suffered 9, BRADTOTHEBONE with 10 and pyromant with 16 in just this season. And my +1 is still not explained unless "someone" would like to claim that I'm lying just to "be right".

[redacted].

You're simply mistaken. Happens when people don't keep personal records.

Edited by McValium, 24 March 2018 - 01:20 PM.
removed nonconstructive

[Der Geisterbaer]

Bombast, on 23 March 2018 - 12:29 PM, said:

You're simply mistaken.

~laugh~ Yeah I totally forgot that one draw that didn't happen ...

Bombast, on 23 March 2018 - 12:29 PM, said:

Happens when people don't keep personal records.

And you know what kind of "personal records" I keep and which I don't?

[Bombast]

Der Geisterbaer, on 23 March 2018 - 12:32 PM, said:

~laugh~ Yeah I totally forgot that one draw that didn't happen ...

And you know what kind of "personal records" I keep and which I don't?

I'm not playing this game anymore. Put up or so on and so forth.

Unrelated, I retroactively apologize to everyone in this thread. Typing on a phone with stupid fingers and I know everything's getting butchered.

[Der Geisterbaer]

Bombast, on 23 March 2018 - 12:35 PM, said:

I'm not playing this game anymore.

~shrug~ That's your prerogative

Bombast, on 23 March 2018 - 12:35 PM, said:

Put up or so on and so forth.

Here's the thing: No matter what I'd put "up or so on and so forth" you'd still be just as "bent" on trusting those numbers as you alleged I'm bent on not being taken serious and directly or indirctly accuse me of either falsifying or just "misremembering" things. I got a reasonable amount of entertainment - for free - so I'll take my spoils for now and let you think whatever you like.

[RustyBolts]

@ Process, I agree that certain players are using it to dismiss other players. in fact it happened during a match when someone told another player that he was ranked higher on Jarls, therefor the other player was **** and didn't know what he was talking about. That was the point I decided to see what the heck Jarls was about. It reminded me of when the Tier system came out and people where claiming that if you weren't Tier 1 or 2 you didn't know what you were talking about. But of course you take a bunch of gamers and give them any kind of stats and the epeen measurements start.

Edited by RustyBolts, 23 March 2018 - 12:51 PM.

[Bombast]

Der Geisterbaer, on 23 March 2018 - 12:47 PM, said:

~shrug~ That's your prerogative



Here's the thing: No matter what I'd put "up or so on and so forth" you'd still be just as "bent" on trusting those numbers as you alleged I'm bent on not being taken serious and directly or indirctly accuse me of either falsifying or just "misremembering" things. I got a reasonable amount of entertainment - for free - so I'll take my spoils for now and let you think whatever you like.

Not true. I think a screenshot of every match you've played from that month, proving you didn't have a mega DC or draw, would go a long way towards proving your position.

But that would be time consuming, and you'd rather bring Flat Earth style argument to the forums. And who doesn't love those guys.

[justcallme A S H]

Bombast, on 23 March 2018 - 12:35 PM, said:

I'm not playing this game anymore.

Sadly most threads with discussion go down this path.

- Someone makes some wildly, inaccurate claim (usually a bottom barrel player)
- Proof, evidence and explaination provided.
- Many lulz
- Person making claim then tries valiantly to change topic/derail
- More lulz
- Person still keeps going, claiming they have proof
- No proof
- Everyone else, with a brain, just becomes frustrated


What I truly do not understand, in the face of proof/evidence, people still cling to their belief.

Edited by justcallme A S H, 23 March 2018 - 03:42 PM.

[Krivvan]

Der Geisterbaer, on 23 March 2018 - 11:30 AM, said:

I already said that I never set foot in Faction Play, so disregard Emerald Taiga as "inexplicable one off" if you wish to do so but tell me again about the quality of the PGI data and the subsequent quality of derived works like Jarl's list.

The Map stats page has always been bugged and has been that way for years. Same with the module stats. The leaderboard stats update correctly. The base stats too.

[Krivvan]

Der Geisterbaer, on 23 March 2018 - 03:08 PM, said:

a ) the proof that leaderboard stats are indeed correct while the rest of the stats are bonkers

You probably weren't around at the time, but the stats were used in applications that I and others have worked on that recorded individual match stats after each match. One application we worked on, MWO Monitor, worked by finding the difference in stats before and after matches after scraping the base stats and mech stats pages. The application is open source and freely available (although I haven't touched it in years and it's deprecated now)[1]. Other applications that performed MWO stat tracking such as MechCollect[2] were also available and to the author's knowledge used the same method.

In the process of validating it after modifying it from the original application[3], 87 matches were played. Our modification to the program consisted of automatically taking screenshots, which wasn't always successful but that in of itself has no bearing on the experiment, and naming the file name with the stats that the program scraped. A comparison was made between the stats shown at the end of the match in-game and the stats scraped by the application from the base and mech stats page represented by the file name. The raw data is available in unaltered form here: https://www.dropbox....IWpaaaYika?dl=0

22 of the matches experienced a failed screenshot capture and the resulting data was ignored. 65 matches had data available both from the stats page application and the in-game interface. In all 65 matches where a comparison could be done, the stats page application generated the same data as was seen in the in-game interface.

65 matches admittedly isn't enough of a sample size for a confidence level of 99%, but over 62 matches without a discrepancy noted should be enough for at least 95% confidence that the application, and therefore the base stat and mech stat tracking pages, were accurate.

Our results suggest that the MWO stats page was accurate for the purposes of tracking individual match statistics. Further work would be required to definitively conclude that the leaderboards stats page is also accurate, but it's reasonable to assume that.

[1] https://github.com/K...itor_KrivvanMod
[2] https://sourceforge....ts/mechcollect/
[3] sourceforge.net/projects/mwomonitor/

On a side note, the program itself took a screenshot based on when the stats page updated. The fact that it got the timing correct so often would also suggest that the stats page was responsive as well as accurate.

Brb, submitting to IEEE T-G and awaiting peer review.

Edited by Krivvan, 23 March 2018 - 04:10 PM.

[Vxheous]

[redacted]
The discrepancy of your +1 game is usually due to a disconnect. If you're disconnected for more than 2 minutes from a game while still alive, the stats will account for the "game played" but not your performance in that match, not W/L, K/D, nor do you get any rewards for any events happening at that point.

Here's the thing though: Jarl's list and MWO leaderboards have identical data. Any missing data is from PGI's end of things, and Jarl's is not attempting to omit any data. There is no intentional tampering on Jarl's compared to MWO leaderboards that somehow makes certain player's stats better than others. Whether I agree with the algorithm of "adjusted match score" vs just using normal average match score and the weighting % of each class is another matter.

Edited by McValium, 24 March 2018 - 01:24 PM.
removed quote since post is removed