Right aging and random talent modifiers for modern day MLB League

[BusterKing]

Originally Posted by SkyDog

Sweet. That's good enough for me. Appreciate the evaluation.

So those last charts are of a result from these settings ?

RATINGS: 80%
CURRENT YEAR: 15%
PREVIOUS YEAR: 4%
2 YEARS AGO: 1%

Sabermetric Lineups

BATTER AGING: .250
BATTER DEV: 1.000
PITCHER AGING: .375
PITCHER DEV: .875

INJURIES: OOTP Classic
TALENT CHANGE RANDOMNESS: 67

[luger]

Originally Posted by BusterKing

So those last charts are of a result from these settings ?

RATINGS: 80%
CURRENT YEAR: 15%
PREVIOUS YEAR: 4%
2 YEARS AGO: 1%

Sabermetric Lineups

BATTER AGING: .250
BATTER DEV: 1.000
PITCHER AGING: .375
PITCHER DEV: .875

INJURIES: OOTP Classic
TALENT CHANGE RANDOMNESS: 67

Indeed they are.

[phightin]

Eiskrap is there any way you can measure the top 100-150 hitters and breakdown the age demographic?

[SunDevil]

Thank you for all the work. But I'm concerned that there aren't any measurements for pitching. The numbers look good for hitting and plate appearances, but Skydog could you provide some pitching data for Eiskrap to graph?

[Mets Man]

I also wanted to say thank you to everybody for their hard work.

One major concern I have about putting the AI Evaluation consideration of Ratings so high is that when it comes time for contract offers, players who put up bad numbers will still be asking for tons of $$$$$$$

I always preferred for everybody (myself and the AI) to emphasize stats over ratings (at least most recent stats). I'm not crazy about the idea of having ratings so highly emphasized. I understand your reasoning behind it, but 80% seems high.

[Ben E Lou]

Originally Posted by Mets Man

I also wanted to say thank you to everybody for their hard work.

One major concern I have about putting the AI Evaluation consideration of Ratings so high is that when it comes time for contract offers, players who put up bad numbers will still be asking for tons of $$$$$$$

I always preferred for everybody (myself and the AI) to emphasize stats over ratings (at least most recent stats). I'm not crazy about the idea of having ratings so highly emphasized. I understand your reasoning behind it, but 80% seems high.

That gets to the core of how you play the game, really. For me, I have zero interest in evaluating a player's talent. I turn scouts off, go with 100% accuracy, and 1-100 ratings scales everywhere. I'm interesting in putting together the best team I can and watching how it performs, and to put the AI on more-equal footing, I really need the AI teams to have a very high emphasis on ratings, so we're doing the same things. The good news for you, though, is that it doesn't make a very big difference in terms of the aging/dev curves, so I suspect you could use the settings I have suggested, tweak the AI evaluation to your liking, and get pretty similar results.

[luger]

Originally Posted by phightin

Eiskrap is there any way you can measure the top 100-150 hitters and breakdown the age demographic?

While I've been tweaking settings in my own fictional league, I've been comparing the age of the top 100 in VORP to real-life so I think this would be handy too.

Originally Posted by SunDevil

The numbers look good for hitting and plate appearances, but Skydog could you provide some pitching data for Eiskrap to graph?

Agreed. In that same vein, the top 50 or 75 SP's compared to real life would be handy.

[luger]

I decided to check it for myself. Here's a graph comparing the top 100 in VORP from that file to MLB's top 100 in batting VORP from 2011. It's not showing multiple years like the previous person did but it at least shows that this data is close to matching up to current real life. The development portion seems spot-on but there might be a tad too many older guys in comparison to current MLB (granted, just one year of data).



Edit: Added a second chart to show age ranges to better visual how there's a few too many talented old players in that particular year versus 2011. Not saying it's a huge flaw in these settings necessarily though.

[Ben E Lou]

Pitchers: http://www.younglifenorthdekalb.com/ootp/pitchers.zip

[SunDevil]

Eiskra,

When you get a chance is there a thread that you could point to that explains how you are able to graph the data? Also, I'm interested in how you coordinated the data out of OOTP and compared it to the data out of the Lahman database.

Thanks.

[luger]

If you add the Data Analysis add-on to Excel and choose the Histogram function, it will put the nice format so you can run the Scatter line chart to get those graphs. Lehman's database can be downloaded in a CSV format so you can open it in Excel as well.

That's how I did my chart above and I assume Eiskrap used a similar procedure though I'm sure he'll chime in if he did something different.

[Eiskrap]

I've attached my template here if you want to do something similar.

All the Lahman data is already in there so all you have to do is export the csv files from the game. Filter them to just just be the Major league data and years you are interested in. Paste the players data in to the Master tab, career batting stats in to Batting Stats and career pitching stats into Pitching Stats. The last thing is to make sure the formula in columns AF:AG of the batting stats and AY:AZ of the pitching stats reach the bottom of the pasted data.

After its calculated the graphs should be on the final two tabs. Should only take 5 minutes and is a useful check.

[SunDevil]

Thanks Eiskrap. But I must be missing something from your post. I do not want this to turn into an excel explanation as I consider myself pretty good, as I understand and can do pivot tables, but this statement is not in your template that you provided:

The last thing is to make sure the formula in columns AF:AG of the batting stats and AY:AZ of the pitching stats reach the bottom of the pasted data.

Both of those tabs (Batting Stats and Pitching Stats) in your attachment are completely empty as there are no formulas in those columns.

I will list out my steps that I took, so maybe you can point out what I did wrong.

1. Downloaded attached template
2. Turned on editing and saved as (not a 97-03 excel format because of loss of features) but instead as Microsoft office 2010 excel document.
3. The Batting Stats, Pitching Stats and Master tabs are completely blank.
4. I go to my league and select the Export Data to CSV files.
5. I then open the Players.csv file, create a new tab. I copy and paste into the Masters tab in your spreadsheet.
6. I then open players_career_batting_stats.csv and I again create a new tab. I then filter the League Id to only be of value 100 (MLB League Id). I then do a date between 2012 through 2032 for the years I want of the MLB specific player career batting stats. I finally filter the split id for only value of 1 (as I do not include left/right splits) and I then copy and paste this data into the new tab created. I then copy and paste the data out of the new tab and paste it into the Batting Stats tab of your spreadsheet.

I will stop here so it does not get much longer, but what did I do wrong or what is the next step specifically.

After doing all the above, this is what I see in Batting Charts tab.

[Cras]

perhaps you copied and pasted formulae rather than data? Right click->Paste Special->Paste Values

[SunDevil]

Nope.

[SunDevil]

So I figured out what I did wrong and it was a stupid mistake. I was copying over the columns AF and AG and wiping out the age and year of birth formulas. Everything showed up. Thank you again for sharing your template.

[Mets Man]

I'm wondering where do you guys find the statistics to put into the chart template? Like where do you gather the data from within OOTP?

[SunDevil]

Originally Posted by Mets Man

I also wanted to say thank you to everybody for their hard work.

One major concern I have about putting the AI Evaluation consideration of Ratings so high is that when it comes time for contract offers, players who put up bad numbers will still be asking for tons of $$$$$$$

I always preferred for everybody (myself and the AI) to emphasize stats over ratings (at least most recent stats). I'm not crazy about the idea of having ratings so highly emphasized. I understand your reasoning behind it, but 80% seems high.

Just wanted to add:

First test simmed 20 years till the end of the 2032 season
25/25/25/25 AI evaluation ratings : highest contract $52 million a year. This was a long term contract and it was the in the last year. The contract started at $26 million and escalated in the 7th year (last year) of the contract to $52 million.

Top 5
$52 million
$50 million
$48 million
$46 million
$40 million

Second test simmed 20 years till the end of the 2032 season
80/15/4/1 AI evaluation ratings: highest contract $40 million a year. This was a long terms contract and it was the last year.

Top 5
$40 million
$36 million
$36 million
$32 million
$32 million

I was not expecting this result because I agree with Mets man that I thought the ai would be giving out bigger contracts with the rating so heavily focused. Not really conclusive with it only being 2 20 year tests but I thought I would report what I saw.

[SunDevil]

Originally Posted by Mets Man

I'm wondering where do you guys find the statistics to put into the chart template? Like where do you gather the data from within OOTP?

Setup & Options->League Setup->Export data to CSV files

This button is located in the bottom right hand corner of the League Setup screen. Then you want three files:

players.csv
Import all the players as some MLB players over the years end up in the minors when they retire, but you need their info so copy everyone over into the template.

players_career_batting_stats.csv
players_career_pitching_stats.csv
Filter them for the league id of 100 (MLB) and the split id of 1 (which is combined left and right handed) and then you need to filter the years. For my two tests I did a filter on the years by doing a between statement of 2012 and 2032.

[Mets Man]

Originally Posted by Eiskrap

I've attached my template here if you want to do something similar.

All the Lahman data is already in there so all you have to do is export the csv files from the game. Filter them to just just be the Major league data and years you are interested in. Paste the players data in to the Master tab, career batting stats in to Batting Stats and career pitching stats into Pitching Stats. The last thing is to make sure the formula in columns AF:AG of the batting stats and AY:AZ of the pitching stats reach the bottom of the pasted data.

After its calculated the graphs should be on the final two tabs. Should only take 5 minutes and is a useful check.

I was curious, how do you get VORP stats to go through the graph?

I only see the ABs, AVG, OPS for batters and IPs, ERA for pitchers.