How IP Gain Has Really Changed (Chart Included)

First Riot Post
1234511 ... 19
Comment below rating threshold, click here to show it.

Subdue

This user has referred a friend to League of Legends, click for more information

Senior Member

03-31-2011

There has been a lot of controversy as to whether or not Riot's analysis of IP gains has been accurate. A lot of us feel that IP gain over time has in fact been reduced. To illustrate this point, I've created an excel sheet that maps out IP gains based on average game length. I don't have Riot's data on overall average game length, so I can't provide that information for you, but through the excel sheet you will be able to find your own average IP per minute and how it's changed since the IP system was changed.

To use the table, simply change the green box labeled "Game Time" to the appropriate value (in minutes). The rest will adjust accordingly.

What the table will tell you based on your average game time:
- Your IP from a win under both systems.
- Your IP from a loss under both systems.
- Your IP per game average based on 50% win/loss.
- Your IP per minute based on 50% win/loss.

A few things to note:

1. I don't have Riot's exact IP gain formula, but I do have a very close approximation, thanks to wildfire393:

Quote:
Originally Posted by wildfire393 View Post
Old IP System: 120 IP for a win at or before 25 minutes, minus 1 IP per minute to a minimum of 102 at 43 minutes.
63 IP for a loss at or before 25 minutes, plus a half an IP per minute to a maximum of 72 IP at 43 minutes.

New IP System: 76 IP for a win at or before 25 minutes, plus a little more than 2 IP per minute to a maximum of 144 IP at 55 minutes.
52 Ip for a loss at or before 25 minutes, plus 1.4 IP per minute to a maximum of 94 IP at 55 minutes.
2. I don't know how Riot adjusts for games under the 25 minute mark. As of now I assume that the IP value continues down from the base at the same rate it rises at.

3. The cross over point is at about 38 minutes. Games that end in less than 38 minutes give less IP/minute under the new system, and games that end after more than 38 minutes give more IP/minute under the new system.

4. The table does not account for the first win of the day bonus, but that shouldn't be a significant factor since it has not changed between systems.

5. The table only accounts for in game time. Time spent in queue and in champ select are not included, and if they were, would skew the information further towards the old system, before game length was shortened.

Edit (4/1/11):
I've edited the Excel sheet to add a new measurement: Weighted IP/Min. Weighted IP/Min is an attempt to achieve a closer approximation to IP/Min by including 5 minutes in the calculation to account for time spent in queue, in champ select, and in loading. In practice, normal games typically take slightly less (probably closer to 3 or 4 minutes for most people, and ranked games take slightly more (probably closer to 7 or 8 minutes for most people. I chose to estimate closer to the normal game time since more people play normals than ranked.


Comment below rating threshold, click here to show it.

Taoist Magic

Senior Member

03-31-2011

nice

i believe 38 min is a little below the average game time actually...but since when did LoL become about grinding for IP instead of enjoying the game?

Now turn it into a web document and have it hosted somewhere.


Comment below rating threshold, click here to show it.

Uccisore

Senior Member

03-31-2011

I thought the discussion in your other thread was plenty fine. To reiterate- you cannot calculate IP per minute, per hour, per day, or per any other amount of time by comparing how much IP you get in a game in the new system vs. the old one. You need to know how long games are, and how long the typical person plays for.

In other words, your IP/minute figure is a lie- all it gives is IP/minute in that game, which is irrelevant.


Comment below rating threshold, click here to show it.

Subdue

This user has referred a friend to League of Legends, click for more information

Senior Member

03-31-2011

Quote:
Originally Posted by Uccisore View Post
I thought the discussion in your other thread was plenty fine. To reiterate- you cannot calculate IP per minute, per hour, per day, or per any other amount of time by comparing how much IP you get in a game in the new system vs. the old one. You need to know how long games are, and how long the typical person plays for.
Actually, this is incorrect. If you actually took a look at the chart, you'd see that it takes game time as a variable. Given the length of a game it is possible to calculate the actual IP gains for that particular game under both systems. It's then just a matter of dividing the gains by the game time you started with to find IP/minute.

Also, while it's true that we have no way of knowing how long the average game is (since only Riot really knows that), it is entirely possible for you to take a sampling of your own games to see how long YOUR average game is, and then see for yourself how the change has affected you.


Comment below rating threshold, click here to show it.

Phreak

Senior Recruiter

03-31-2011
1 of 1 Riot Posts

In short, you're mostly correct. We took a "snap shot" of the average game time, took the relevant IP gains, and said, "Okay, this is going to be the IP gain per minute for League of Legends going forward." We constructed a system that would produce those values (plus accounting for time in champion select, loading, etc., so it has a "base" value in addition to the "per-minute" IP).

As has been said by others, some users will end up losing IP in the long run because their games were shorter than the average. Some users will end up gaining IP in the long run because they games were longer on average. This is unfortunate for those of you who ended up getting hurt by this system change, so I'm sorry for that.

The reason we are using the new system is to allow for a better balance between maps. Instead of forcing -22% on Treeline for average shorter game times, we have a system that makes everything fair regardless of averages. This is a much cleaner way to go about making all maps accessible without having to constantly research and update multiplier values for IP/XP gains.


Comment below rating threshold, click here to show it.

Uccisore

Senior Member

03-31-2011

Quote:
Originally Posted by Dominatius Maxim View Post
Actually, this is incorrect. If you actually took a look at the chart, you'd see that it takes game time as a variable. Given the length of a game it is possible to calculate the actual IP gains for that particular game under both systems. It's then just a matter of dividing the gains by the game time you started with to find IP/minute.
Ip/minute IN THAT GAME. That's not going to be your IP/minute for, say, three hours of play, or a week of play. Like I said in your other thread, since you can't buy anything with 1 game's worth if IP (and never could), IP/minute in one game is irrelevant to our concerns. What we really want to know is if it takes more or less time to earn, say, 6300 IP. Your table does not help us with that.


Comment below rating threshold, click here to show it.

EcksTeaSea

This user has referred a friend to League of Legends, click for more information

Senior Member

03-31-2011

Quote:
Originally Posted by Phreak View Post
In short, you're mostly correct. We took a "snap shot" of the average game time, took the relevant IP gains, and said, "Okay, this is going to be the IP gain per minute for League of Legends going forward." We constructed a system that would produce those values (plus accounting for time in champion select, loading, etc., so it has a "base" value in addition to the "per-minute" IP).

As has been said by others, some users will end up losing IP in the long run because their games were shorter than the average. Some users will end up gaining IP in the long run because they games were longer on average. This is unfortunate for those of you who ended up getting hurt by this system change, so I'm sorry for that.
Then why not just change it back where everyone was happy? My games mostly run at 20 minutes now and with you guys making the games go faster, its all bout the shorter game now.


Comment below rating threshold, click here to show it.

Desperate Orphan

This user has referred a friend to League of Legends, click for more information

Senior Member

03-31-2011

Quote:
Originally Posted by Uccisore View Post
Ip/minute IN THAT GAME. That's not going to be your IP/minute for, say, three hours of play, or a week of play. Like I said in your other thread, since you can't buy anything with 1 game's worth if IP (and never could), IP/minute in one game is irrelevant to our concerns. What we really want to know is if it takes more or less time to earn, say, 6300 IP. Your table does not help us with that.
this.


Comment below rating threshold, click here to show it.

Uccisore

Senior Member

03-31-2011

Quote:
Originally Posted by Dominatius Maxim View Post
it is entirely possible for you to take a sampling of your own games to see how long YOUR average game is, and then see for yourself how the change has affected you.
Only if you can somehow compare it to what your average game time used to be before the changes.


Comment below rating threshold, click here to show it.

Orinoko

Senior Member

03-31-2011

Quote:
Originally Posted by Phreak View Post
In short, you're mostly correct. We took a "snap shot" of the average game time, took the relevant IP gains, and said, "Okay, this is going to be the IP gain per minute for League of Legends going forward." We constructed a system that would produce those values (plus accounting for time in champion select, loading, etc., so it has a "base" value in addition to the "per-minute" IP).

As has been said by others, some users will end up losing IP in the long run because their games were shorter than the average. Some users will end up gaining IP in the long run because they games were longer on average. This is unfortunate for those of you who ended up getting hurt by this system change, so I'm sorry for that.
I really don't understand why you wouldn't just change it back to the original IP gains to keep EVERYONE happy.


1234511 ... 19