Relative Ratings suggest

[saturn2187]

Good Morning


When using Relative Ratings, what should we look for when deciding whether to promote?

I'm in an online league, which until recently didn't have any roster limits. Roster limits finally did just go into effect, and in order to get the teams down to size, I had to self impose age limits. This helped cut the total numbers, but it does nothing for helping shepherd the best prospects up the chain.


I've been waiting until they show 50s in the level to which they are headed, which I know is far too late because they aren't getting to AAA until 26 .


What does everybody else do?

[saturn2187]

Bah humbug

[NoOne]

if you have relative turned on, then 50% of your scale is likely 'average' for that league as it is currently constituted, if i understand how it works, correctly. that doesn't necessarily mean they are average for that league level - Just relative to players in the league, which will change each year.

when they are above average relative to players in that league, they are probably approaching average for the next step up right? probably ready for a promotion.... but still depends on things... even more things to worry about when you leave it on "relative".

so, it comes down to opinion on whether you fast-track, let them simmer alot, or somewhere in between. do you want them ~average, a little below average, or ready to hit the ground running with good success at the next level?

i've never used the setting, i prefer that the ratings always mean the same thing to me... i dont' have to guess if the league is "better" than average and making the player look worse than they really are -- high levels of talent will make an 'average' player be rated lower than average relative to the gluttony of talent in that league. the simple fact that the AI so poorly places amatuers is enough reason to avoid this... too many players playing at the wrong level!

some people swear by it, but it empirically makes it more difficult to know what you have... the player is less precisely defined... you need to apply some subjective evaluations (or do real work for a 'video game') by eye of the league's talent and how it is different from any other year to understand what a 50/100 means in any given year, lol. i'd bet on a 5-10% swing in ratings for any year... so a 10% drop can occur and you don't know if it's real, tcr or inaccuracy... 1 more possibility than not using relative.

if you have a very small scale, it becomes less silly. some scales are so small that it's difficult to differentiate player at lower levels... 20-80 or 100pt scale is plenty to see, though... 0-5 or 1-10 may benefit from a relative scale, i am not sure... but leaving the door open.

(you can guess i don't like relative scale nor relative to position - same reasoning)

[saturn2187]

Originally Posted by NoOne

if you have relative turned on, then 50% of your scale is likely 'average' for that league, if i understand how it works, correctly.

when they are above average for that league, they are probably approaching average for the next step up right?

so, it comes down to opinion on whether you fast-track, let them simmer alot, or somewhere in between.

i've never used the setting, i prefer that the ratings always mean the same thing to me... i dont' have to guess if the league is "better" than average and making the player look worse than they really are -- high levels of talent will make an 'average' player be rated lower than average relative to the gluttony of talent in that league.

some people swear by it, but it empirically makes it more difficult to know what you have... the player is less accurately defined... you need to apply some subjective evaluations by eye of the league's talent and how it is different from any other year to understand what a 50/100 means in any given year, lol.

if you have a very small scale, it becomes less silly. some scales are so small that it's difficult to differentiate player at lower levels... 20-80 or 100pt scale is plenty to see, though... 0-5 or 1-10 may benefit from a relative scale, i am not sure... but leaving the door open.

(you can guess i don't like relative scale nor relative to position - same reasoning)

Thanks for the reply, I think....

I'm in an Online League where this change was voted upon by the league, even though after it went into effect there was widespread confusion about what had just happened. Anyway, there's no going back now so I am just trying to get used to the new system.

[NoOne]

so, i'd operate under the assumption that when they are above 50% of contact scale at that level, they are candidates for promotion based on ratings...

it will depend on a couple things... does it shift the scale so that 50 is average? you'll know right away when "20"/100 eye becomes 50/100 in rookie league. if that's the case, 50% will be a good starting point for each rating. otherwise you'll have to use common sense as to where the middle is for each rating. turn on and off and see exactly what it is doing as a whole and work up from there.

depends on scouting accuracy - you may also use statitistics... keep using those for promote/demote as you have always done. the stats won't change. if you normally promote a guy in rookie that's hitting .350, continue to do that. also, use that stuff to augment what you are learning about "relative to league" changes in ratings.

the only caveat for ratings is that you have to make some guess as to whether the league is significantly high or low in talent that year... most years are probably not a concern... maybe a sharp chagne in ratings will be the best omen for this occurence.

___ Workaround for seeing through "relative"

i know!!! you can keep track of a few "filler" players - ones that will never make the majors and you don't care about leaving at various levels year to year - also they can't be changing much as far as their ratings (not developing is best). this will work better in AA/AAA when less development occurs with crappy minor league lifers, but there are always players with maxed out development that can't go higher... keep them instaed of cutting for this reason. That A-ball guy with really bad, but fully developed ratings all off a sudden becomes a keeper in order to understand what the ratings mean year-to-year. it's not like they will play at all, and if so you can force them to the bench one way or another.

anyway, if yu have a known quantity - one that isn't developing much more or has stagnated .... leave them at that level and you can tell by how his ratings change year to year what "relative" is -- up or down etc... more than one would be safe due to scouting inaccuracy.

you'll have a solid guess as to whether the league is up or down, currently, based on how that player looks year to year. at least year-to-year initially and with enough familiarity evenatually the bigger picture too.

if you keep scouting reports, this will be easy - as long as the "relative" setting does not change past scouting ratings, retroactively - i doubt it, but somethign to check. othewise, just screen shot a few rosters and compare those players that didn't move the next year and likely didn't change ratings much as far as development or TCR effects. you will see the shift in ratings, if any.