Does server ranking effect drop rates?

MrAlBobo

L13: Stunning Member
Feb 20, 2008
1,059
219
Earlier today I joined a server I haven't played on in a long time, a 24/7 dustbowl server. While on this server I found 3 items over the period of an hour. This struck me as very strange as ive been playing exclusively on the tf2maps servers for about 2 weeks now, and during the last week ive found one item.

3 over an hour vs 1 over 15-20 hours (a guess based on my total playtime)
something was clearly up...or my luck was just really bad then it got good for an hour

Later on I remembered a blog update on the teamfortress site:
http://www.teamfortress.com/post.php?id=2338

This basically showed how valve was tracking servers and rating them based on how satisfied their players were. They did mention at the end of the post that they did intend to try and use this/ expand on it, but nothing was ever said.

# Each time a player connects to a server, it loses 15 points
# For each minute the player stays on the server, it earns 1 point (up to a max of 45 points per player)

This is the criteria listed for rating the servers. Now...think about this for a moment, based on this having a server that is frequently populated (so one or two people don't join and leave from boredom), one that caters to what the majority want to play, and one where people stay connected for long periods of time would be good for points.

Earlier on I mentioned that 24/7 dustbowl server, this server is almost always populated, it does precisely what people who go to it want, a game of dustbowl, and its scheduled for hour long rounds (when the map changes). So this server is bound to be rated very highly. And, on this highly rated server I managed to find 3 items in an hour, along with seeing at least 10 other people get items in that same period.

Now take a look at the tf2maps servers, they remain empty most of the time, they exclusively run customs, something known (in tf2 anyways) for high disconnect rates. We also, instead of having a fresh group of people every hour or so like most server that can fill themselves, we have the same people there most of the time that would not continue earning server points. Incidentally, as mentioned earlier ive got one item from our servers in the past week, and ive only seen 3-4 other people get items that I can recall.

Then, once looking at these examples, think of a third implication, the idle server (don't bitch about idling here). People who join want to sit there for long periods of time. Very few people will join an idle server and leave in under an hour as that defeats the purpose. So while not gaining points rapidly, idle servers will still rise as they will not really be losing points.


As it stands this is just a guess of mine, but with my small sample it does serve to explain a lot of things that didn't really make sense. So...anyone that notices anything like this should post so I can see if my examples aren't just fluke.
 

Fragimus_Max

L3: Member
Jan 13, 2008
146
57
That's pretty clever, and I have to say, dude... it makes A LOT of sense.

I think I got an item drop on the TF2Maps server within four sessions, and then I went to some 24/7 last night and got like two drops in about two hours or less (actually, I think it was Dustbowl too!)

I also deleted items: an Ubersaw and a Huntsman... and guess what I got back?

Another Ubersaw and my first Jar of Karate. I found it strange that, since I had all the Medic items, I got a dupe - and I didn't have Jarate, but removed the Huntsman from inventory and got what I needed to fill a void for that class.

I rarely make deletions, but I also remember deleting a Sandman and finding a Bonk soon after... which was another item I did not have at the time. I'm VERY curious as to what would happen if I deleted two high level Ambassadors... would that get me something like the fedora?

Might be worth a try. Something deep within tells me (gas?) that the levels *are* actually worth something. Why would they even bother adding that if it had nothing to do with any type of functionality - be it visible or hidden.

Also, Idlers are infamous for deleting items when the backpack get too full. Could this somehow be tied into the whole trading system we've seen the video of?
 
Last edited:

Nineaxis

Quack Doctor
aa
May 19, 2008
1,767
2,820
I asked Robin for you.

No, drop rates aren’t affected by server stats.



One thing to be aware of though: TF shows items drops to players when they enter spectator mode after the freezecam ends. Some instant respawn servers run plugins that prevent players ever reaching spectator mode, so players are never shown their drops. While they’re earning drops as normal, it appears that no-one on the server is doing so. As soon as a player dies on a server without that plugin, they’ll be shown all the drops they’d accumulated. It’s possible this is the cause of the suspicion.



Robin.
 

MrAlBobo

L13: Stunning Member
Feb 20, 2008
1,059
219
I asked Robin for you.

As I said in chat I still question if he would say this provided they were actually using this system...it really seems like something that would be exploited, or just used as a tool to screw over servers that you hate

and the respawn part is irrelevant

I forgot server rankings even existed. Can you even view the rankings?

As far as I know you can't
 

A Boojum Snark

Toraipoddodezain Mazahabado
aa
Nov 2, 2007
4,775
7,669
It's all meaningless without a large scale test. The nature of randomness is to be random.
Also, I have always felt the "per player" thing implies per session. Not that once you are on a server for 45 minutes you will never ever provide that server with points again, that would be dumb.
 

MrAlBobo

L13: Stunning Member
Feb 20, 2008
1,059
219
It's all meaningless without a large scale test. The nature of randomness is to be random.
Also, I have always felt the "per player" thing implies per session. Not that once you are on a server for 45 minutes you will never ever provide that server with points again, that would be dumb.

it is per session...and I thought I wrote it with that in mind :p

also I want the large scale testing...I wouldn't post it otherwise
 

Terr

Cranky Coder
aa
Jul 31, 2009
1,590
410
I got so frustrated with people in the TF2 forums saying: "omfg bad idea valve" that I outlined an efficient* one-demotion-per-person-per-server-per--day implementation in SQL with code snippets. I think it got pruned. Darn.

Anyway, it's been painful to watch the "They said 'maggots'! SOLDR UPDAET NEXT!!111" folks get ahold of anecdotal item-drop information and then breathlessly draw wholly unsupported conclusions. A solid statistics class should be required for high-school graduation, I swear...

*Insofar as any computer programmer thinks their way is the efficient one.
 

MrAlBobo

L13: Stunning Member
Feb 20, 2008
1,059
219
I got so frustrated with people in the TF2 forums saying: "omfg bad idea valve" that I outlined an efficient* one-demotion-per-person-per-server-per--day implementation in SQL with code snippets. I think it got pruned. Darn.

Anyway, it's been painful to watch the "They said 'maggots'! SOLDR UPDAET NEXT!!111" folks get ahold of anecdotal item-drop information and then breathlessly draw wholly unsupported conclusions. A solid statistics class should be required for high-school graduation, I swear...

*Insofar as any computer programmer thinks their way is the efficient one.

honestly...half of this I don't even see the relevance

the other half seems to be referring to my "guess" regarding the drop system on the same level as the hordes of retards that read into every little detail of what valve says and instantly assume its correct

mralbobo said:
As it stands this is just a guess of mine, but with my small sample it does serve to explain a lot of things that didn't really make sense. So...anyone that notices anything like this should post so I can see if my examples aren't just fluke.

You will notice that I used the word "guess" to describe this, I did not use the word theory, nor even the word hypothesis. Primarily due to my incredibly small sample. This was simply a thought I had, and I decided to post it and see if anyone else noticed anything similar, or had any useful information to disprove it.

That said I got another 2 hours of playtime in on that server mentioned earlier, and found nothing, so I am leaning toward it being a fluke. Though more data is still needed.
 
Last edited:

Terr

Cranky Coder
aa
Jul 31, 2009
1,590
410
I wasn't trying to conflate your carefully-hedged* theory with the Steam Forum madness.

*Non-pejorative, I swear.
 

MrAlBobo

L13: Stunning Member
Feb 20, 2008
1,059
219
I wasn't trying to conflate your carefully-hedged* theory with the Steam Forum madness.

*Non-pejorative, I swear.

alright then...i figured you were either making a comment like I said earlier or saying something completely off topic

apparently i took a shot and missed <_<
 

grazr

Old Man Mutant Ninja Turtle
aa
Mar 4, 2008
5,441
3,814
I play a 24/7 2fort server mostly. It's always populated throughout the 24 hours of a day. between 3am and 6am it can drop as low as 12vs12, in an otherwise 33 player server. But the players tend to take it in 4-5 hour sessions. Atleast the regulars.

I don't notice that many drops for myself, and a reasonable amount of drops get issued between players i suppose. 1 drop for a single player in 40 minutes (if that can be seen as reasonable? But then i've always seen my personal luck of finding drops scewed as i tend to survive up to 25 mins at a time. I could swear living long lives prevents the oppotunity to pick up more items, as as far as i'm aware you can only be awarded 1 item per life. There seems to be a trend whereby bad players seem to get way more items than good players.

That said. Deleting an item on recieval doesn't seem to flag the usual text alert on the HUD/in the server. Could this be effecting our opinion of drop rates? Recieving drops but filtering out duplicates before they hit our backpack?
 

MrAlBobo

L13: Stunning Member
Feb 20, 2008
1,059
219
That said. Deleting an item on recieval doesn't seem to flag the usual text alert on the HUD/in the server. Could this be effecting our opinion of drop rates? Recieving drops but filtering out duplicates before they hit our backpack?

Huh...I actually never thought of that...mostly because ive never had a full backpack but still...interesting