After the baseball season is over, the work begins. Every player is rated by handed using TPS. The ratings are reviewed for quality-control purposes and finalized. When possible, a three year baseline is used, with last season having the most weight (50:25:25 weighting). If less than three seasons of data are available, regression to the mean is applied.
Since last season was only 60 games, fielding calculations were weighted more towards the previous season where possible. If a player has played 162 career games or less and has played more than 1 MLB season, his career splits are used in 2/3 of the calculation.
For long term franchise play, the potential mix is adjusted to divide players into relevant tiers, based on current production and age.
Potential Mix for MLB Position Players
A (90-99) = 16%
B (80-89) = 34%
C (70-79) = 51%
Potential Mix for MLB Pitchers
A (90-99) = 18%
B (80-89) = 36%
C (70-79) = 46%
Hitting and Fielding
Hitting and Fielding
-Contact ratings are based on AVG
-Power ratings are based on ISO and HR%
-Vision ratings are based on K%
-Discipline ratings are based on BB%
-Durability ratings are based on the league average games played per MLB season (102 games) for all position players
-Speed (baserunning ability) is calculated based on a formula that takes into account sprint speed, player position, stolen bases, XBHs, and GIDP rate. Statcast's jump metric is also taken into account for outfielders.
-Fielding ratings (arm accuracy, reaction and fielding ability) are position specific; catchers, infielders and outfielders are all rated differently.
-RDRS (Baseball Info Solutions Defensive Runs Saved Above Average) is factored into the calculations, which makes position specific adjustments. Statcast's OAA (Outs Above Average) is used for outfielders. Statcast's jump data drives the reaction ratings for outfielders.
Pitchers
-Per 9 ratings are based on the per 9 statistics.
-Stamina is based on innings/appearance
Pitch repetoires were built from the ground up using Statcast data via MLB Savant. Pitch types were influencd by relative horizontal and vertical break for each pitch, using the % vs average differentiator:
-Fastballs with more relative horizontal break than vertical break are Running Fastballs
-Sinkers with more relative vertical break than horizontal break remain Sinkers. If a pitcher's relative creer groundball percentage is >=5%, the Sinker stays a Sinker as well. Flyball pitchers are more likely to receive two-seamers as well.
-Changeups with more relative horizontal break than vertical break are Circle Changeups.
-Curveballs (<80 mph) with more relative vertical break than horizontal break are 12-6 Curveballs
-Curveballs (<80 mph) with more relative horizontal brek than vertical break are Sweeping Curveballs.
-Curveballs (>80 mph) wih more relative vertical break than horizonal break are Knucklecurves
-Curveballs (>80 mph) with more relative horizontal break than vertical break are Slurves.
Prospects
For long term franchise play, the potential mix is adjusted to divide players into relevant tiers, based on current production and age.
Potential Mix for MLB Position Players
A (90-99) = 16%
B (80-89) = 34%
C (70-79) = 51%
Potential Mix for MLB Pitchers
A (90-99) = 18%
B (80-89) = 36%
C (70-79) = 46%
The potential system was first changed for MLB 20 to improve long-term balance in franchise. The original methodology behind that is as follows:
|
Quote: |
|
|
|
|
|
|
|
|
|
Minor league players and major league players potentials are rated differently. Pitchers and hitters are also on different potential scales.
When a player reaches the majors, how much they were touted in the minors doesn't matter anymore. What matters is how well they produce. From the player's production, we can assess what their ceiling is.
Prospect ratings and potentials were determined based on:
-Age
-Current Ability (2019 stats/most recent season)
-Highest level of competition
Hitters and pitchers progress a bit differently. Progression for hitters is more random/performance based than pitchers.
For hitters, in order to get a good baseline, prospects from previous years were used as comparrisons. The goal was to look at players retroactively and see if TPS could accurately predict how good they would become. I looked at the 2017 Top 100 Prospects list, since we now have a reasonable idea of how those prospects turned out.
This was helpful because every year, there are superstars on this list and there are busts. My goal was to follow a similar potential spread the MLB rosters have, where there are a few stars, some above average players, and everyday/bench guys.
In designing the potential calculation method, I wanted to prioritize how advanced the hitter's approach was (walk%, whiff%) and how well they could actually hit. The best players in baseball have great plate discipline and limit strikeouts, with the occasional exception. However, they can all hit. Simply put, if a player can't hit, his ceiling is lower and vice versa. The quality of competition and age are then considered to avoid overrating a player's ceiling. One mistake many make in evaluating prospects in rosters is eering on the side of high potentials. When potentials are too high, franchise is unplayable beyond year 2 and there is no balance long-term.
Wander Franco is the consensus #1 prospect in baseball right now. One may look at the reports and say oh, 99 potential for sure. I like Franco, but he has only played in A ball. For players in A, there is an additional potential adjustment. Instead of a 99, TPS gives Franco a 90, which is still superstar level, rather than hall-of-fame/pereniel MVP candidate level (i.e. Trout).
If Franco had played against better competition with a similar performance (AA or AAA) then he would likely be a 99.
For pitchers, it is more likely that a player will achieve his potential rating than hitters in MLBTS, regardless of performance.
This was taken into account and a similar approach was taken to evaluate them.
There are three ceilings for pitchers:
75: #4/#5 starter
80: #3/#2 starter
90: Potential Ace
Their performance against level of competition and how deep they go into games determined their ceiling (potential rating). |
|
|
|
|
|
While the method used to assign potentials to pitchers was not changed, the potentials for hitters were adjusted this year because franchise no longer uses morale to give ratings/OVR boosts or penalties to players. This means that the potentials do not need to be as rigid for hitters.
Based on Prospect Pipeline, our adjusted potential scale for hitting prospects looks like:
Potential-MLB Pipeline Overall Grade
99 - 75+
90 - 60+
85 - 55
80 - 50 or in team's top 10
75 (if in team's top 15) - 45
70 - 45
65 - 40