forums

Miami Baseball Season Projections

View as article
Lance Roffers

Lance Roffers

Junior
Joined
Feb 23, 2018
Messages
1,987
Baseball season is back! Miami will open the season with their annual series versus Rutgers and before things get started for real it’s time for projections. Last year I predicted a record of 34-20 and the Canes fell short of that record with a final showing at 28-26. However, if you look at the projected record for Miami using BaseRuns, they finished with a projected winning percentage of .587, which would result in a record of 32-22, so the quality of the team pretty accurately reflected the projections if not for some bad sequencing luck.

What is BaseRuns?
BaseRuns is a metric that tries to remove “sequencing” from the equation to determine overall strength of a club. Sequencing is simply the order in which events occur and have shown to be quite random in nature. For instance, in a one game sample a team who has two hits and one walk can beat a team with nine hits and five walks if the sequencing allows for it. I. E. a walk, hit, then HR would create three runs for Team A, and Team B may have the more hits and walks, but if the sequencing is spread out over the course of the game they may not score any runs at all. Over the course of a season, these events should even out and create a more realistic impression of what a team should be in quality. If you are into more advanced stats this is metric is sometimes called “Second Order Wins.”

The difference between Miami’s “Expected winning percentage” and their actual winning percentage was fairly large at 0.068 (.587 - .519). In the ACC, the only clubs who had a larger gap between their expected winning percentage and their actual winning percentage were Boston College and Wake Forest, who tied for the largest gap at 0.086. Clemson was the most fortunate club in the ACC, outproducing their expected winning percentage by -0.070 (0.763 to an expected 0.693).

Projections:
Running projections for college players is a less refined exercise than what you’d see at the major league level. Smaller sample sizes, more unequal quality of competition, and park factors all can impact the results of players at the college level that are less impactful at the professional level. I’ve regressed hitting projections towards the mean in Batting Average on Balls In Play (BABIP), BB%, K%, HR%. For pitching, I’ve regressed projections towards the mean in BB%, K%, HR%, BABIP, HBP%.

The main hitting metric is Weighted On-Base Average (wOBA) which attachs linear weights to each offensive outcome and attempts to weight the outcomes into a number that is on the same scale as On-Base Percentage for the league.

The main pitching metric is Fielding Independent Pitching (FIP) which removes defensive ability from things the pitcher is directly responsible for (walks, strikeouts, hit-by-pitch, HR’s). Once a pitch is released, the pitcher no longer has any control on the outcome of the at-bat, it is up to the defense. This number is scaled to match the league average ERA.

Freddy Zamora- 210 AB’s, .319/.402/.438, .344 BABIP, 10.5% BB, 8.9% K, 0.365 wOBA

Tony Jenkins- 199 AB’s, .256/.398/.377, .322 BABIP, 17.1% BB, 19.9% K, 0.348 wOBA

Willy Escala- 196 AB’s, .281/.376/.372, .380 BABIP, 10.9% BB, 24.3% K, 0.332 wOBA

Dylan Cloonan- 193 AB’s, .269/.388/.404, .360 BABIP, 15.4% BB, 23.5% K, 0.351 wOBA

Michael Amditis- 145 AB’s, .269/.329/.331, .328 BABIP, 7.4% BB, 18.4% K, 0.291 wOBA

Adrian Del Castillo- 124 AB’s, .269/.371/.427, .340 BABIP, 10.5% BB, 19.6% K, 0.343 wOBA

Anthony Vilar- 123 AB’s, .236/.309/.301, .315 BABIP, 7.9% BB, 22.3% K, 0.274 wOBA

Chad Crosbie- 111 AB’s, .270/.383/.369, .345 BABIP, 14.2% BB, 20.1% K, 0.336 wOBA

Chet Moore- 110 AB’s, .273/.385/.327, .357 BABIP, 12.8% BB, 19.5% K, 0.325 wOBA

Alex Toral- 102 AB’s, .225/.379/.382, .308 BABIP, 17.4% BB, 28% K, 0.338 wOBA

Isaac Quinones- 98 AB’s, .265/.348/.357, .333 BABIP, 8.9% BB, 20.5% K, 0.313 wOBA

Jordan Lala- 93 AB’s, .258/.352/.323, .333 BABIP, 10.9% BB, 20% K, 0.304 wOBA

Ray Gil- 79 AB’s, .263/.348/.418, .340 BABIP, 10.5% BB, 26.3% K, 0.333 wOBA

JP Gates- 76 AB’s, .263/.348/.408, .305 BABIP, 11.2% BB, 19.1% K, 0.327 wOBA

Gabe Rivera- 67 AB’s, .254/.338/.448, .311 BABIP, 9.1%, 26% K, 0.334 wOBA



Pitchers

Evan McKendry- 96 IP, 3.47 ERA, 2.80 FIP, 31.2% K, 8.1% BB, 23.1% K-BB

Chris McMahon- 80 IP, 3.94 ERA, 3.61 FIP, 20.7% K, 7.5% BB, 13.2% K-BB

Greg Veliz- 69 IP, 3.39 ERA, 3.50 FIP, 25.3% K, 12.5% BB, 12.8% K-BB

Brian Van Belle- 56 IP, 4.47 ERA, 4.59 FIP, 16.9% K, 9.3% BB, 7.6% K-BB

Daniel Federman- 44 IP, 2.86 ERA, 3.04 FIP, 24.5% K, 8.2% BB, 16.3% K-BB

Slade Cecconi- 41 IP, 3.73 ERA, 3.77 FIP, 23.5% K, 11.8%, 11.8% K-BB

JP Gates- 39 IP, 3.00 ERA, 3.65 FIP, 24.4% K, 10.7% K, 13.7% K-BB

Mark Mixon- 23 IP, 3.47 ERA, 3.56 FIP, 31.4% K, 10.5% BB, 20.9% K-BB

Tyler Keysor- 22 IP, 3.27 ERA, 3.93 FIP, 23.6% K, 12.4% BB, 11.2% K-BB

Jeremy Cook- 13IP, 4.85 ERA, 6.01 FIP, 25% K, 21.4% BB, 3.6% K-BB

Albert Maury- 13 IP, 4.85 ERA, 4.94 FIP, 17.6% K, 11.8% BB, 5.9% K-BB

Daniel Rivero- 10 IP, 6.30 ERA, 6.47 FIP, 16.7% K, 18.8% BB, -2.1% K-BB



Overall Projections
I know many of you skip to the end, so I’ll wrap it up from a team perspective here. Personal numbers are all regressed to account for several different outcomes, but the total team numbers will be fairly similar to what we will end the year with from a quantity standpoint.

Overall, I have Miami hitting .268/.370/.378 as a team with a 0.331 wOBA. For comparison, last year Miami hit .257/.356/.358 with a 0.317 wOBA so I do have them improving quite a bit. The average team in the ACC last year hit .268/.372/.406 for a 0.340 wOBA. When you factor in park factors for many of the clubs hitting in better hitting parks for their home games, I have Miami projected as a roughly average offensive club this year.

On the pitching side I have Miami being a standout with a team ERA of 3.71, a 3.71 FIP, 24.1% K, 10.5% BB, 13.6% K-BB. For comparison, last year Miami was at 3.74 ERA, 3.83 FIP, 23% K, 14.4% BB, 8.6% K-BB. In the ACC, the average team had an ERA of 4.13, a 4.13 FIP, 21% K, 16.5% BB, 4.6% K-BB rate.

Using the two metrics together, you can compile an expected winning percentage using BaseRuns. Miami checks in with a winning percentage of .647, which means I am projecting a record of 37-19.

Because I’m a madman, I built an RPI calculator and input the schedule for this year into the calculator. While it uses last year’s RPI numbers for the opponents, generally enough opponents will improve/decline at a similar rate that the numbers are a nice proxy for the next season. Plugging in 37 wins for Miami in a way that makes sense given who/where they play teams, you get to a projected RPI of .5719. Last year, that RPI had this cluster of teams around that number (Jacksonville, NC State, Coastal Carolina, Missouri State) and would’ve been number 24. NC State and Coastal Carolina hosted Regionals, while Jacksonville was a two-seed in Gainesville and Missouri State was a three-seed in the Ole Miss Regional.

This projection will clearly get Miami back into the NCAA tournament and put them squarely on the hosting bubble. I look forward to an excellent season and getting Miami baseball back to the tournament where it belongs.
 
Advertisement
Advertisement
Lance Roffers said:
Baseball season is back! Miami will open the season with their annual series versus Rutgers and before things get started for real it’s time for projections. Last year I predicted a record of 34-20 and the Canes fell short of that record with a final showing at 28-26. However, if you look at the projected record for Miami using BaseRuns, they finished with a projected winning percentage of .587, which would result in a record of 32-22, so the quality of the team pretty accurately reflected the projections if not for some bad sequencing luck.

What is BaseRuns?
BaseRuns is a metric that tries to remove “sequencing” from the equation to determine overall strength of a club. Sequencing is simply the order in which events occur and have shown to be quite random in nature. For instance, in a one game sample a team who has two hits and one walk can beat a team with nine hits and five walks if the sequencing allows for it. I. E. a walk, hit, then HR would create three runs for Team A, and Team B may have the more hits and walks, but if the sequencing is spread out over the course of the game they may not score any runs at all. Over the course of a season, these events should even out and create a more realistic impression of what a team should be in quality. If you are into more advanced stats this is metric is sometimes called “Second Order Wins.”

The difference between Miami’s “Expected winning percentage” and their actual winning percentage was fairly large at 0.068 (.587 - .519). In the ACC, the only clubs who had a larger gap between their expected winning percentage and their actual winning percentage were Boston College and Wake Forest, who tied for the largest gap at 0.086. Clemson was the most fortunate club in the ACC, outproducing their expected winning percentage by -0.070 (0.763 to an expected 0.693).

Projections:
Running projections for college players is a less refined exercise than what you’d see at the major league level. Smaller sample sizes, more unequal quality of competition, and park factors all can impact the results of players at the college level that are less impactful at the professional level. I’ve regressed hitting projections towards the mean in Batting Average on Balls In Play (BABIP), BB%, K%, HR%. For pitching, I’ve regressed projections towards the mean in BB%, K%, HR%, BABIP, HBP%.

The main hitting metric is Weighted On-Base Average (wOBA) which attachs linear weights to each offensive outcome and attempts to weight the outcomes into a number that is on the same scale as On-Base Percentage for the league.

The main pitching metric is Fielding Independent Pitching (FIP) which removes defensive ability from things the pitcher is directly responsible for (walks, strikeouts, hit-by-pitch, HR’s). Once a pitch is released, the pitcher no longer has any control on the outcome of the at-bat, it is up to the defense. This number is scaled to match the league average ERA.

Freddy Zamora- 210 AB’s, .319/.402/.438, .344 BABIP, 10.5% BB, 8.9% K, 0.365 wOBA

Tony Jenkins- 199 AB’s, .256/.398/.377, .322 BABIP, 17.1% BB, 19.9% K, 0.348 wOBA

Willy Escala- 196 AB’s, .281/.376/.372, .380 BABIP, 10.9% BB, 24.3% K, 0.332 wOBA

Dylan Cloonan- 193 AB’s, .269/.388/.404, .360 BABIP, 15.4% BB, 23.5% K, 0.351 wOBA

Michael Amditis- 145 AB’s, .269/.329/.331, .328 BABIP, 7.4% BB, 18.4% K, 0.291 wOBA

Adrian Del Castillo- 124 AB’s, .269/.371/.427, .340 BABIP, 10.5% BB, 19.6% K, 0.343 wOBA

Anthony Vilar- 123 AB’s, .236/.309/.301, .315 BABIP, 7.9% BB, 22.3% K, 0.274 wOBA

Chad Crosbie- 111 AB’s, .270/.383/.369, .345 BABIP, 14.2% BB, 20.1% K, 0.336 wOBA

Chet Moore- 110 AB’s, .273/.385/.327, .357 BABIP, 12.8% BB, 19.5% K, 0.325 wOBA

Alex Toral- 102 AB’s, .225/.379/.382, .308 BABIP, 17.4% BB, 28% K, 0.338 wOBA

Isaac Quinones- 98 AB’s, .265/.348/.357, .333 BABIP, 8.9% BB, 20.5% K, 0.313 wOBA

Jordan Lala- 93 AB’s, .258/.352/.323, .333 BABIP, 10.9% BB, 20% K, 0.304 wOBA

Ray Gil- 79 AB’s, .263/.348/.418, .340 BABIP, 10.5% BB, 26.3% K, 0.333 wOBA

JP Gates- 76 AB’s, .263/.348/.408, .305 BABIP, 11.2% BB, 19.1% K, 0.327 wOBA

Gabe Rivera- 67 AB’s, .254/.338/.448, .311 BABIP, 9.1%, 26% K, 0.334 wOBA



Pitchers

Evan McKendry- 96 IP, 3.47 ERA, 2.80 FIP, 31.2% K, 8.1% BB, 23.1% K-BB

Chris McMahon- 80 IP, 3.94 ERA, 3.61 FIP, 20.7% K, 7.5% BB, 13.2% K-BB

Greg Veliz- 69 IP, 3.39 ERA, 3.50 FIP, 25.3% K, 12.5% BB, 12.8% K-BB

Brian Van Belle- 56 IP, 4.47 ERA, 4.59 FIP, 16.9% K, 9.3% BB, 7.6% K-BB

Daniel Federman- 44 IP, 2.86 ERA, 3.04 FIP, 24.5% K, 8.2% BB, 16.3% K-BB

Slade Cecconi- 41 IP, 3.73 ERA, 3.77 FIP, 23.5% K, 11.8%, 11.8% K-BB

JP Gates- 39 IP, 3.00 ERA, 3.65 FIP, 24.4% K, 10.7% K, 13.7% K-BB

Mark Mixon- 23 IP, 3.47 ERA, 3.56 FIP, 31.4% K, 10.5% BB, 20.9% K-BB

Tyler Keysor- 22 IP, 3.27 ERA, 3.93 FIP, 23.6% K, 12.4% BB, 11.2% K-BB

Jeremy Cook- 13IP, 4.85 ERA, 6.01 FIP, 25% K, 21.4% BB, 3.6% K-BB

Albert Maury- 13 IP, 4.85 ERA, 4.94 FIP, 17.6% K, 11.8% BB, 5.9% K-BB

Daniel Rivero- 10 IP, 6.30 ERA, 6.47 FIP, 16.7% K, 18.8% BB, -2.1% K-BB



Overall Projections
I know many of you skip to the end, so I’ll wrap it up from a team perspective here. Personal numbers are all regressed to account for several different outcomes, but the total team numbers will be fairly similar to what we will end the year with from a quantity standpoint.

Overall, I have Miami hitting .268/.370/.378 as a team with a 0.331 wOBA. For comparison, last year Miami hit .257/.356/.358 with a 0.317 wOBA so I do have them improving quite a bit. The average team in the ACC last year hit .268/.372/.406 for a 0.340 wOBA. When you factor in park factors for many of the clubs hitting in better hitting parks for their home games, I have Miami projected as a roughly average offensive club this year.

On the pitching side I have Miami being a standout with a team ERA of 3.71, a 3.71 FIP, 24.1% K, 10.5% BB, 13.6% K-BB. For comparison, last year Miami was at 3.74 ERA, 3.83 FIP, 23% K, 14.4% BB, 8.6% K-BB. In the ACC, the average team had an ERA of 4.13, a 4.13 FIP, 21% K, 16.5% BB, 4.6% K-BB rate.

Using the two metrics together, you can compile an expected winning percentage using BaseRuns. Miami checks in with a winning percentage of .647, which means I am projecting a record of 37-19.

Because I’m a madman, I built an RPI calculator and input the schedule for this year into the calculator. While it uses last year’s RPI numbers for the opponents, generally enough opponents will improve/decline at a similar rate that the numbers are a nice proxy for the next season. Plugging in 37 wins for Miami in a way that makes sense given who/where they play teams, you get to a projected RPI of .5719. Last year, that RPI had this cluster of teams around that number (Jacksonville, NC State, Coastal Carolina, Missouri State) and would’ve been number 24. NC State and Coastal Carolina hosted Regionals, while Jacksonville was a two-seed in Gainesville and Missouri State was a three-seed in the Ole Miss Regional.

This projection will clearly get Miami back into the NCAA tournament and put them squarely on the hosting bubble. I look forward to an excellent season and getting Miami baseball back to the tournament where it belongs.
Click to expand...

Lance, thanks for all the analysis and time put in. You seem to enjoy it, but it still has to be a lot of work researching all the individual stats.

Question, - is there a metric that measures base-running aggressiveness/effectiveness, as I expect Gino to be much more aggressive in attempting to manufacture runs this year, e.g. SB/SBA, first to third w less than two outs, squeeze bunts, etc?

I have watched MLB for the last several years almost every night and continue to be amazed w how clueless and careless they are on the base paths. I would think it would be as meaningful a statistic as defensive runs saved.
 
80441
 
Advertisement
BP Joe said:
Lance, thanks for all the analysis and time put in. You seem to enjoy it, but it still has to be a lot of work researching all the individual stats.

Question, - is there a metric that measures base-running aggressiveness/effectiveness, as I expect Gino to be much more aggressive in attempting to manufacture runs this year, e.g. SB/SBA, first to third w less than two outs, squeeze bunts, etc?

I have watched MLB for the last several years almost every night and continue to be amazed w how clueless and careless they are on the base paths. I would think it would be as meaningful a statistic as defensive runs saved.
Click to expand...

Yes, this is actually a metric tracked by Fangraphs and it’s called simply, “Baserunning.”

It’s difficult to calculate at the collegiate level and I think there’s some guesswork involved with the public models.

I do calculate a metric entitled “Weighted Stolen Base” which seeks to give a “run value” on the stolen base attempts. The Major League shows you need to have at least a 70% success rate to be adding value so I’m using their formula, but honestly, with the higher frequency of runs and extra base hits, you truly probably need to be closer to 75% to be making it worthwhile in many situations.

Game theory calculations can do the math on each situation to tell you what the best move is from a statistical standpoint, but I still believe there is a place for gut feel in this game as well.
 
Lance Roffers said:
Baseball season is back! Miami will open the season with their annual series versus Rutgers and before things get started for real it’s time for projections. Last year I predicted a record of 34-20 and the Canes fell short of that record with a final showing at 28-26. However, if you look at the projected record for Miami using BaseRuns, they finished with a projected winning percentage of .587, which would result in a record of 32-22, so the quality of the team pretty accurately reflected the projections if not for some bad sequencing luck.

What is BaseRuns?
BaseRuns is a metric that tries to remove “sequencing” from the equation to determine overall strength of a club. Sequencing is simply the order in which events occur and have shown to be quite random in nature. For instance, in a one game sample a team who has two hits and one walk can beat a team with nine hits and five walks if the sequencing allows for it. I. E. a walk, hit, then HR would create three runs for Team A, and Team B may have the more hits and walks, but if the sequencing is spread out over the course of the game they may not score any runs at all. Over the course of a season, these events should even out and create a more realistic impression of what a team should be in quality. If you are into more advanced stats this is metric is sometimes called “Second Order Wins.”

The difference between Miami’s “Expected winning percentage” and their actual winning percentage was fairly large at 0.068 (.587 - .519). In the ACC, the only clubs who had a larger gap between their expected winning percentage and their actual winning percentage were Boston College and Wake Forest, who tied for the largest gap at 0.086. Clemson was the most fortunate club in the ACC, outproducing their expected winning percentage by -0.070 (0.763 to an expected 0.693).

Projections:
Running projections for college players is a less refined exercise than what you’d see at the major league level. Smaller sample sizes, more unequal quality of competition, and park factors all can impact the results of players at the college level that are less impactful at the professional level. I’ve regressed hitting projections towards the mean in Batting Average on Balls In Play (BABIP), BB%, K%, HR%. For pitching, I’ve regressed projections towards the mean in BB%, K%, HR%, BABIP, HBP%.

The main hitting metric is Weighted On-Base Average (wOBA) which attachs linear weights to each offensive outcome and attempts to weight the outcomes into a number that is on the same scale as On-Base Percentage for the league.

The main pitching metric is Fielding Independent Pitching (FIP) which removes defensive ability from things the pitcher is directly responsible for (walks, strikeouts, hit-by-pitch, HR’s). Once a pitch is released, the pitcher no longer has any control on the outcome of the at-bat, it is up to the defense. This number is scaled to match the league average ERA.

Freddy Zamora- 210 AB’s, .319/.402/.438, .344 BABIP, 10.5% BB, 8.9% K, 0.365 wOBA

Tony Jenkins- 199 AB’s, .256/.398/.377, .322 BABIP, 17.1% BB, 19.9% K, 0.348 wOBA

Willy Escala- 196 AB’s, .281/.376/.372, .380 BABIP, 10.9% BB, 24.3% K, 0.332 wOBA

Dylan Cloonan- 193 AB’s, .269/.388/.404, .360 BABIP, 15.4% BB, 23.5% K, 0.351 wOBA

Michael Amditis- 145 AB’s, .269/.329/.331, .328 BABIP, 7.4% BB, 18.4% K, 0.291 wOBA

Adrian Del Castillo- 124 AB’s, .269/.371/.427, .340 BABIP, 10.5% BB, 19.6% K, 0.343 wOBA

Anthony Vilar- 123 AB’s, .236/.309/.301, .315 BABIP, 7.9% BB, 22.3% K, 0.274 wOBA

Chad Crosbie- 111 AB’s, .270/.383/.369, .345 BABIP, 14.2% BB, 20.1% K, 0.336 wOBA

Chet Moore- 110 AB’s, .273/.385/.327, .357 BABIP, 12.8% BB, 19.5% K, 0.325 wOBA

Alex Toral- 102 AB’s, .225/.379/.382, .308 BABIP, 17.4% BB, 28% K, 0.338 wOBA

Isaac Quinones- 98 AB’s, .265/.348/.357, .333 BABIP, 8.9% BB, 20.5% K, 0.313 wOBA

Jordan Lala- 93 AB’s, .258/.352/.323, .333 BABIP, 10.9% BB, 20% K, 0.304 wOBA

Ray Gil- 79 AB’s, .263/.348/.418, .340 BABIP, 10.5% BB, 26.3% K, 0.333 wOBA

JP Gates- 76 AB’s, .263/.348/.408, .305 BABIP, 11.2% BB, 19.1% K, 0.327 wOBA

Gabe Rivera- 67 AB’s, .254/.338/.448, .311 BABIP, 9.1%, 26% K, 0.334 wOBA



Pitchers

Evan McKendry- 96 IP, 3.47 ERA, 2.80 FIP, 31.2% K, 8.1% BB, 23.1% K-BB

Chris McMahon- 80 IP, 3.94 ERA, 3.61 FIP, 20.7% K, 7.5% BB, 13.2% K-BB

Greg Veliz- 69 IP, 3.39 ERA, 3.50 FIP, 25.3% K, 12.5% BB, 12.8% K-BB

Brian Van Belle- 56 IP, 4.47 ERA, 4.59 FIP, 16.9% K, 9.3% BB, 7.6% K-BB

Daniel Federman- 44 IP, 2.86 ERA, 3.04 FIP, 24.5% K, 8.2% BB, 16.3% K-BB

Slade Cecconi- 41 IP, 3.73 ERA, 3.77 FIP, 23.5% K, 11.8%, 11.8% K-BB

JP Gates- 39 IP, 3.00 ERA, 3.65 FIP, 24.4% K, 10.7% K, 13.7% K-BB

Mark Mixon- 23 IP, 3.47 ERA, 3.56 FIP, 31.4% K, 10.5% BB, 20.9% K-BB

Tyler Keysor- 22 IP, 3.27 ERA, 3.93 FIP, 23.6% K, 12.4% BB, 11.2% K-BB

Jeremy Cook- 13IP, 4.85 ERA, 6.01 FIP, 25% K, 21.4% BB, 3.6% K-BB

Albert Maury- 13 IP, 4.85 ERA, 4.94 FIP, 17.6% K, 11.8% BB, 5.9% K-BB

Daniel Rivero- 10 IP, 6.30 ERA, 6.47 FIP, 16.7% K, 18.8% BB, -2.1% K-BB



Overall Projections
I know many of you skip to the end, so I’ll wrap it up from a team perspective here. Personal numbers are all regressed to account for several different outcomes, but the total team numbers will be fairly similar to what we will end the year with from a quantity standpoint.

Overall, I have Miami hitting .268/.370/.378 as a team with a 0.331 wOBA. For comparison, last year Miami hit .257/.356/.358 with a 0.317 wOBA so I do have them improving quite a bit. The average team in the ACC last year hit .268/.372/.406 for a 0.340 wOBA. When you factor in park factors for many of the clubs hitting in better hitting parks for their home games, I have Miami projected as a roughly average offensive club this year.

On the pitching side I have Miami being a standout with a team ERA of 3.71, a 3.71 FIP, 24.1% K, 10.5% BB, 13.6% K-BB. For comparison, last year Miami was at 3.74 ERA, 3.83 FIP, 23% K, 14.4% BB, 8.6% K-BB. In the ACC, the average team had an ERA of 4.13, a 4.13 FIP, 21% K, 16.5% BB, 4.6% K-BB rate.

Using the two metrics together, you can compile an expected winning percentage using BaseRuns. Miami checks in with a winning percentage of .647, which means I am projecting a record of 37-19.

Because I’m a madman, I built an RPI calculator and input the schedule for this year into the calculator. While it uses last year’s RPI numbers for the opponents, generally enough opponents will improve/decline at a similar rate that the numbers are a nice proxy for the next season. Plugging in 37 wins for Miami in a way that makes sense given who/where they play teams, you get to a projected RPI of .5719. Last year, that RPI had this cluster of teams around that number (Jacksonville, NC State, Coastal Carolina, Missouri State) and would’ve been number 24. NC State and Coastal Carolina hosted Regionals, while Jacksonville was a two-seed in Gainesville and Missouri State was a three-seed in the Ole Miss Regional.

This projection will clearly get Miami back into the NCAA tournament and put them squarely on the hosting bubble. I look forward to an excellent season and getting Miami baseball back to the tournament where it belongs.
Click to expand...
Outstanding Lance! I love statistical analysis, but you put one on me with your work.
 
Lance,

Awesome stuff. I'm assuming for returners you used previous year's results, but how did you calculate projections for Freshman with no real stats to go off of? High school stats can be misleading.

Thanks for the read!
 
Advertisement
rokulika said:
Lance, any guess what the batting order might be?
Click to expand...

At this point, I’d assume it’s fairly similar to what they did in the alumni game outside of maybe Del Castillo being DH and Cloonan starting instead of Chet Moore (if it’s a righty starter).
 
PGrady24 said:
Lance,

Awesome stuff. I'm assuming for returners you used previous year's results, but how did you calculate projections for Freshman with no real stats to go off of? High school stats can be misleading.

Thanks for the read!
Click to expand...

It’s not easy. You start with a freshman baseline for the ACC and then adjust for recruiting ranking, information you get from coaches from practice/fall, use information about the type of hitter they were in college (plate discipline, swing traits), physical build, experience watching them.

You’re regressing everything towards a 50th percentile projection, which is why projection systems end up with like one or zero MLB players having 120 runs, RBI’s, etc. even though we know several will hit those marks. It’s just the people hitting those marks are hitting their 70-80th percentile projections.

For college freshmen there is a lot of educated guessing and regression off precedence.

Last year I had Alex Toral projected pretty low compared to college websites because his HS performance didn’t dictate him being a freshman of year type or hitting for power people expected. I was way too light on Quinones and Jenkins though.
 
Lance Roffers said:
Because I’m a madman, I built an RPI calculator and input the schedule for this year into the calculator. While it uses last year’s RPI numbers for the opponents, generally enough opponents will improve/decline at a similar rate that the numbers are a nice proxy for the next season. Plugging in 37 wins for Miami in a way that makes sense given who/where they play teams, you get to a projected RPI of .5719. Last year, that RPI had this cluster of teams around that number (Jacksonville, NC State, Coastal Carolina, Missouri State) and would’ve been number 24. NC State and Coastal Carolina hosted Regionals, while Jacksonville was a two-seed in Gainesville and Missouri State was a three-seed in the Ole Miss Regional.

This projection will clearly get Miami back into the NCAA tournament and put them squarely on the hosting bubble. I look forward to an excellent season and getting Miami baseball back to the tournament where it belongs.
Click to expand...

As you may have figured out, I am also an avid RPI follower and also run projections as we get enough data in the system. One thing to be careful about when looking at RPI: conference finish is a lot more important to the committee than ordinal rankings. We have seen teams with top 25 RPIs completely miss the tournament.

The most important things for us: stop losing to bad non-conference teams, and finish in the top 5 or 6 in the ACC.
 
Advertisement
Jaromir Jagr said:
As you may have figured out, I am also an avid RPI follower and also run projections as we get enough data in the system. One thing to be careful about when looking at RPI: conference finish is a lot more important to the committee than ordinal rankings. We have seen teams with top 25 RPIs completely miss the tournament.

The most important things for us: stop losing to bad non-conference teams, and finish in the top 5 or 6 in the ACC.
Click to expand...

This ^^^^^^^^. It has been so frustrating the past few seasons losing to teams we should be dominating. We fix this and we are a top 15 team and hosting a regional.
 
Lance Roffers said:
Baseball season is back! Miami will open the season with their annual series versus Rutgers and before things get started for real it’s time for projections. Last year I predicted a record of 34-20 and the Canes fell short of that record with a final showing at 28-26. However, if you look at the projected record for Miami using BaseRuns, they finished with a projected winning percentage of .587, which would result in a record of 32-22, so the quality of the team pretty accurately reflected the projections if not for some bad sequencing luck.

What is BaseRuns?
BaseRuns is a metric that tries to remove “sequencing” from the equation to determine overall strength of a club. Sequencing is simply the order in which events occur and have shown to be quite random in nature. For instance, in a one game sample a team who has two hits and one walk can beat a team with nine hits and five walks if the sequencing allows for it. I. E. a walk, hit, then HR would create three runs for Team A, and Team B may have the more hits and walks, but if the sequencing is spread out over the course of the game they may not score any runs at all. Over the course of a season, these events should even out and create a more realistic impression of what a team should be in quality. If you are into more advanced stats this is metric is sometimes called “Second Order Wins.”

The difference between Miami’s “Expected winning percentage” and their actual winning percentage was fairly large at 0.068 (.587 - .519). In the ACC, the only clubs who had a larger gap between their expected winning percentage and their actual winning percentage were Boston College and Wake Forest, who tied for the largest gap at 0.086. Clemson was the most fortunate club in the ACC, outproducing their expected winning percentage by -0.070 (0.763 to an expected 0.693).

Projections:
Running projections for college players is a less refined exercise than what you’d see at the major league level. Smaller sample sizes, more unequal quality of competition, and park factors all can impact the results of players at the college level that are less impactful at the professional level. I’ve regressed hitting projections towards the mean in Batting Average on Balls In Play (BABIP), BB%, K%, HR%. For pitching, I’ve regressed projections towards the mean in BB%, K%, HR%, BABIP, HBP%.

The main hitting metric is Weighted On-Base Average (wOBA) which attachs linear weights to each offensive outcome and attempts to weight the outcomes into a number that is on the same scale as On-Base Percentage for the league.

The main pitching metric is Fielding Independent Pitching (FIP) which removes defensive ability from things the pitcher is directly responsible for (walks, strikeouts, hit-by-pitch, HR’s). Once a pitch is released, the pitcher no longer has any control on the outcome of the at-bat, it is up to the defense. This number is scaled to match the league average ERA.

Freddy Zamora- 210 AB’s, .319/.402/.438, .344 BABIP, 10.5% BB, 8.9% K, 0.365 wOBA

Tony Jenkins- 199 AB’s, .256/.398/.377, .322 BABIP, 17.1% BB, 19.9% K, 0.348 wOBA

Willy Escala- 196 AB’s, .281/.376/.372, .380 BABIP, 10.9% BB, 24.3% K, 0.332 wOBA

Dylan Cloonan- 193 AB’s, .269/.388/.404, .360 BABIP, 15.4% BB, 23.5% K, 0.351 wOBA

Michael Amditis- 145 AB’s, .269/.329/.331, .328 BABIP, 7.4% BB, 18.4% K, 0.291 wOBA

Adrian Del Castillo- 124 AB’s, .269/.371/.427, .340 BABIP, 10.5% BB, 19.6% K, 0.343 wOBA

Anthony Vilar- 123 AB’s, .236/.309/.301, .315 BABIP, 7.9% BB, 22.3% K, 0.274 wOBA

Chad Crosbie- 111 AB’s, .270/.383/.369, .345 BABIP, 14.2% BB, 20.1% K, 0.336 wOBA

Chet Moore- 110 AB’s, .273/.385/.327, .357 BABIP, 12.8% BB, 19.5% K, 0.325 wOBA

Alex Toral- 102 AB’s, .225/.379/.382, .308 BABIP, 17.4% BB, 28% K, 0.338 wOBA

Isaac Quinones- 98 AB’s, .265/.348/.357, .333 BABIP, 8.9% BB, 20.5% K, 0.313 wOBA

Jordan Lala- 93 AB’s, .258/.352/.323, .333 BABIP, 10.9% BB, 20% K, 0.304 wOBA

Ray Gil- 79 AB’s, .263/.348/.418, .340 BABIP, 10.5% BB, 26.3% K, 0.333 wOBA

JP Gates- 76 AB’s, .263/.348/.408, .305 BABIP, 11.2% BB, 19.1% K, 0.327 wOBA

Gabe Rivera- 67 AB’s, .254/.338/.448, .311 BABIP, 9.1%, 26% K, 0.334 wOBA



Pitchers

Evan McKendry- 96 IP, 3.47 ERA, 2.80 FIP, 31.2% K, 8.1% BB, 23.1% K-BB

Chris McMahon- 80 IP, 3.94 ERA, 3.61 FIP, 20.7% K, 7.5% BB, 13.2% K-BB

Greg Veliz- 69 IP, 3.39 ERA, 3.50 FIP, 25.3% K, 12.5% BB, 12.8% K-BB

Brian Van Belle- 56 IP, 4.47 ERA, 4.59 FIP, 16.9% K, 9.3% BB, 7.6% K-BB

Daniel Federman- 44 IP, 2.86 ERA, 3.04 FIP, 24.5% K, 8.2% BB, 16.3% K-BB

Slade Cecconi- 41 IP, 3.73 ERA, 3.77 FIP, 23.5% K, 11.8%, 11.8% K-BB

JP Gates- 39 IP, 3.00 ERA, 3.65 FIP, 24.4% K, 10.7% K, 13.7% K-BB

Mark Mixon- 23 IP, 3.47 ERA, 3.56 FIP, 31.4% K, 10.5% BB, 20.9% K-BB

Tyler Keysor- 22 IP, 3.27 ERA, 3.93 FIP, 23.6% K, 12.4% BB, 11.2% K-BB

Jeremy Cook- 13IP, 4.85 ERA, 6.01 FIP, 25% K, 21.4% BB, 3.6% K-BB

Albert Maury- 13 IP, 4.85 ERA, 4.94 FIP, 17.6% K, 11.8% BB, 5.9% K-BB

Daniel Rivero- 10 IP, 6.30 ERA, 6.47 FIP, 16.7% K, 18.8% BB, -2.1% K-BB



Overall Projections
I know many of you skip to the end, so I’ll wrap it up from a team perspective here. Personal numbers are all regressed to account for several different outcomes, but the total team numbers will be fairly similar to what we will end the year with from a quantity standpoint.

Overall, I have Miami hitting .268/.370/.378 as a team with a 0.331 wOBA. For comparison, last year Miami hit .257/.356/.358 with a 0.317 wOBA so I do have them improving quite a bit. The average team in the ACC last year hit .268/.372/.406 for a 0.340 wOBA. When you factor in park factors for many of the clubs hitting in better hitting parks for their home games, I have Miami projected as a roughly average offensive club this year.

On the pitching side I have Miami being a standout with a team ERA of 3.71, a 3.71 FIP, 24.1% K, 10.5% BB, 13.6% K-BB. For comparison, last year Miami was at 3.74 ERA, 3.83 FIP, 23% K, 14.4% BB, 8.6% K-BB. In the ACC, the average team had an ERA of 4.13, a 4.13 FIP, 21% K, 16.5% BB, 4.6% K-BB rate.

Using the two metrics together, you can compile an expected winning percentage using BaseRuns. Miami checks in with a winning percentage of .647, which means I am projecting a record of 37-19.

Because I’m a madman, I built an RPI calculator and input the schedule for this year into the calculator. While it uses last year’s RPI numbers for the opponents, generally enough opponents will improve/decline at a similar rate that the numbers are a nice proxy for the next season. Plugging in 37 wins for Miami in a way that makes sense given who/where they play teams, you get to a projected RPI of .5719. Last year, that RPI had this cluster of teams around that number (Jacksonville, NC State, Coastal Carolina, Missouri State) and would’ve been number 24. NC State and Coastal Carolina hosted Regionals, while Jacksonville was a two-seed in Gainesville and Missouri State was a three-seed in the Ole Miss Regional.

This projection will clearly get Miami back into the NCAA tournament and put them squarely on the hosting bubble. I look forward to an excellent season and getting Miami baseball back to the tournament where it belongs.
Click to expand...
I seriously just joined so that I could tell you that this is impressive! Hats off to you sir!
 
Lance Roffers said:
Baseball season is back! Miami will open the season with their annual series versus Rutgers and before things get started for real it’s time for projections. Last year I predicted a record of 34-20 and the Canes fell short of that record with a final showing at 28-26. However, if you look at the projected record for Miami using BaseRuns, they finished with a projected winning percentage of .587, which would result in a record of 32-22, so the quality of the team pretty accurately reflected the projections if not for some bad sequencing luck.

What is BaseRuns?
BaseRuns is a metric that tries to remove “sequencing” from the equation to determine overall strength of a club. Sequencing is simply the order in which events occur and have shown to be quite random in nature. For instance, in a one game sample a team who has two hits and one walk can beat a team with nine hits and five walks if the sequencing allows for it. I. E. a walk, hit, then HR would create three runs for Team A, and Team B may have the more hits and walks, but if the sequencing is spread out over the course of the game they may not score any runs at all. Over the course of a season, these events should even out and create a more realistic impression of what a team should be in quality. If you are into more advanced stats this is metric is sometimes called “Second Order Wins.”

The difference between Miami’s “Expected winning percentage” and their actual winning percentage was fairly large at 0.068 (.587 - .519). In the ACC, the only clubs who had a larger gap between their expected winning percentage and their actual winning percentage were Boston College and Wake Forest, who tied for the largest gap at 0.086. Clemson was the most fortunate club in the ACC, outproducing their expected winning percentage by -0.070 (0.763 to an expected 0.693).

Projections:
Running projections for college players is a less refined exercise than what you’d see at the major league level. Smaller sample sizes, more unequal quality of competition, and park factors all can impact the results of players at the college level that are less impactful at the professional level. I’ve regressed hitting projections towards the mean in Batting Average on Balls In Play (BABIP), BB%, K%, HR%. For pitching, I’ve regressed projections towards the mean in BB%, K%, HR%, BABIP, HBP%.

The main hitting metric is Weighted On-Base Average (wOBA) which attachs linear weights to each offensive outcome and attempts to weight the outcomes into a number that is on the same scale as On-Base Percentage for the league.

The main pitching metric is Fielding Independent Pitching (FIP) which removes defensive ability from things the pitcher is directly responsible for (walks, strikeouts, hit-by-pitch, HR’s). Once a pitch is released, the pitcher no longer has any control on the outcome of the at-bat, it is up to the defense. This number is scaled to match the league average ERA.

Freddy Zamora- 210 AB’s, .319/.402/.438, .344 BABIP, 10.5% BB, 8.9% K, 0.365 wOBA

Tony Jenkins- 199 AB’s, .256/.398/.377, .322 BABIP, 17.1% BB, 19.9% K, 0.348 wOBA

Willy Escala- 196 AB’s, .281/.376/.372, .380 BABIP, 10.9% BB, 24.3% K, 0.332 wOBA

Dylan Cloonan- 193 AB’s, .269/.388/.404, .360 BABIP, 15.4% BB, 23.5% K, 0.351 wOBA

Michael Amditis- 145 AB’s, .269/.329/.331, .328 BABIP, 7.4% BB, 18.4% K, 0.291 wOBA

Adrian Del Castillo- 124 AB’s, .269/.371/.427, .340 BABIP, 10.5% BB, 19.6% K, 0.343 wOBA

Anthony Vilar- 123 AB’s, .236/.309/.301, .315 BABIP, 7.9% BB, 22.3% K, 0.274 wOBA

Chad Crosbie- 111 AB’s, .270/.383/.369, .345 BABIP, 14.2% BB, 20.1% K, 0.336 wOBA

Chet Moore- 110 AB’s, .273/.385/.327, .357 BABIP, 12.8% BB, 19.5% K, 0.325 wOBA

Alex Toral- 102 AB’s, .225/.379/.382, .308 BABIP, 17.4% BB, 28% K, 0.338 wOBA

Isaac Quinones- 98 AB’s, .265/.348/.357, .333 BABIP, 8.9% BB, 20.5% K, 0.313 wOBA

Jordan Lala- 93 AB’s, .258/.352/.323, .333 BABIP, 10.9% BB, 20% K, 0.304 wOBA

Ray Gil- 79 AB’s, .263/.348/.418, .340 BABIP, 10.5% BB, 26.3% K, 0.333 wOBA

JP Gates- 76 AB’s, .263/.348/.408, .305 BABIP, 11.2% BB, 19.1% K, 0.327 wOBA

Gabe Rivera- 67 AB’s, .254/.338/.448, .311 BABIP, 9.1%, 26% K, 0.334 wOBA



Pitchers

Evan McKendry- 96 IP, 3.47 ERA, 2.80 FIP, 31.2% K, 8.1% BB, 23.1% K-BB

Chris McMahon- 80 IP, 3.94 ERA, 3.61 FIP, 20.7% K, 7.5% BB, 13.2% K-BB

Greg Veliz- 69 IP, 3.39 ERA, 3.50 FIP, 25.3% K, 12.5% BB, 12.8% K-BB

Brian Van Belle- 56 IP, 4.47 ERA, 4.59 FIP, 16.9% K, 9.3% BB, 7.6% K-BB

Daniel Federman- 44 IP, 2.86 ERA, 3.04 FIP, 24.5% K, 8.2% BB, 16.3% K-BB

Slade Cecconi- 41 IP, 3.73 ERA, 3.77 FIP, 23.5% K, 11.8%, 11.8% K-BB

JP Gates- 39 IP, 3.00 ERA, 3.65 FIP, 24.4% K, 10.7% K, 13.7% K-BB

Mark Mixon- 23 IP, 3.47 ERA, 3.56 FIP, 31.4% K, 10.5% BB, 20.9% K-BB

Tyler Keysor- 22 IP, 3.27 ERA, 3.93 FIP, 23.6% K, 12.4% BB, 11.2% K-BB

Jeremy Cook- 13IP, 4.85 ERA, 6.01 FIP, 25% K, 21.4% BB, 3.6% K-BB

Albert Maury- 13 IP, 4.85 ERA, 4.94 FIP, 17.6% K, 11.8% BB, 5.9% K-BB

Daniel Rivero- 10 IP, 6.30 ERA, 6.47 FIP, 16.7% K, 18.8% BB, -2.1% K-BB



Overall Projections
I know many of you skip to the end, so I’ll wrap it up from a team perspective here. Personal numbers are all regressed to account for several different outcomes, but the total team numbers will be fairly similar to what we will end the year with from a quantity standpoint.

Overall, I have Miami hitting .268/.370/.378 as a team with a 0.331 wOBA. For comparison, last year Miami hit .257/.356/.358 with a 0.317 wOBA so I do have them improving quite a bit. The average team in the ACC last year hit .268/.372/.406 for a 0.340 wOBA. When you factor in park factors for many of the clubs hitting in better hitting parks for their home games, I have Miami projected as a roughly average offensive club this year.

On the pitching side I have Miami being a standout with a team ERA of 3.71, a 3.71 FIP, 24.1% K, 10.5% BB, 13.6% K-BB. For comparison, last year Miami was at 3.74 ERA, 3.83 FIP, 23% K, 14.4% BB, 8.6% K-BB. In the ACC, the average team had an ERA of 4.13, a 4.13 FIP, 21% K, 16.5% BB, 4.6% K-BB rate.

Using the two metrics together, you can compile an expected winning percentage using BaseRuns. Miami checks in with a winning percentage of .647, which means I am projecting a record of 37-19.

Because I’m a madman, I built an RPI calculator and input the schedule for this year into the calculator. While it uses last year’s RPI numbers for the opponents, generally enough opponents will improve/decline at a similar rate that the numbers are a nice proxy for the next season. Plugging in 37 wins for Miami in a way that makes sense given who/where they play teams, you get to a projected RPI of .5719. Last year, that RPI had this cluster of teams around that number (Jacksonville, NC State, Coastal Carolina, Missouri State) and would’ve been number 24. NC State and Coastal Carolina hosted Regionals, while Jacksonville was a two-seed in Gainesville and Missouri State was a three-seed in the Ole Miss Regional.

This projection will clearly get Miami back into the NCAA tournament and put them squarely on the hosting bubble. I look forward to an excellent season and getting Miami baseball back to the tournament where it belongs.
Click to expand...



Accuracy bump.

Sitting at 35 wins with 5 to go.
 
Advertisement
Back
Top