Thursday, May 31, 2018

Games of the Day

The Nationals travel to Atlanta as a battle for first place in the NL East begins. Tanner Roark and Sean Newcomb pitch game one of the four game series. Roark comes in with a WHIP of 0.99, as he allowed 65 hits plus walks in 65 1/3 innings. He has a great history against the Braves, 7-3 with a 2.77 ERA, including a great start in early April. Newcomb is striking out and walking batters at about the same rate as his rookie year, but he cut his home runs rate nearly in half. That helped reduce his ERA by about one and a half runs.

Aaron Nola pulls a tough assignment for the Phillies as he faces the Dodgers and the return of Clayton Kershaw. It’s been such a good year for pitchers that Nola’s 2.27 ERA ranks 10th in the majors. He only allowed four home run in 71 1/3 innings. Kershaw allowed seven home runs in 44 innings before the injury. Three of those home runs came off the bats of left-handed batters.

Enjoy!



from baseballmusings.com https://ift.tt/2LOmSzB

Beat the Streak Picks

Here are the top picks my programs produced for use in Beat the Streak. This post mostly explains the ideas behind the calculations. In addition, this post shows tests on the Neural Network (NN). This post discusses an NN that includes the ballpark. I updated the models, and the results of those tests are here.

For 2018, I am just going to publish the Log5 hit averages and the NN probabilities with parks factored in. I am keeping track of the results here. I added a graph that gives a visual representation of the probability and success each day.

I have been asked to expand the list to the top 25 players for an econometric project.

First, the Log5 Method picks:

0.330 — Jose Altuve batting against Drew Pomeranz.
0.324 — Jean Segura batting against Mike Minor.
0.294 — Ronald Torreyes batting against Andrew Cashner.
0.292 — Andrelton Simmons batting against Ryan N Carpenter.
0.288 — Yulieski Gurriel batting against Drew Pomeranz.
0.287 — Manny Machado batting against Sonny Gray.
0.285 — Michael Brantley batting against Jake Odorizzi.
0.284 — Matt Kemp batting against Aaron Nola.
0.284 — Miguel Andujar batting against Andrew Cashner.
0.282 — Eddie Rosario batting against Shane Bieber.
0.280 — Adam Jones batting against Sonny Gray.
0.279 — J.T. Realmuto batting against Jordan Lyles.
0.278 — Wilson Ramos batting against Daniel Mengden.
0.276 — Nick Castellanos batting against Andrew Heaney.
0.276 — Gleyber Torres batting against Andrew Cashner.
0.274 — Asdrubal Cabrera batting against Jose Quintana.
0.274 — Starlin Castro batting against Jordan Lyles.
0.272 — Matt M Duffy batting against Daniel Mengden.
0.272 — George Springer batting against Drew Pomeranz.
0.270 — Odubel Herrera batting against Clayton Kershaw.
0.270 — Albert Almora batting against Seth Lugo.
0.268 — Starling Marte batting against Jack Flaherty.
0.266 — Corey Dickerson batting against Jack Flaherty.
0.265 — Austin Romine batting against Andrew Cashner.
0.265 — Mookie Betts batting against Lance McCullers.
0.264 — Jose Pirela batting against Wei-Yin Chen.

I listed 26 here because I’m not sure of the status of Mookie Betts. The Red Sox may not want to play a man short against the Astros. Altuve is 7 for 17 against Pomeranz in the regular season with two strikeouts. He’s two for two in the post season as well.

Here is how the NN with Park ranks the players:

0.330, 0.761 — Jose Altuve batting against Drew Pomeranz.
0.324, 0.751 — Jean Segura batting against Mike Minor.
0.285, 0.729 — Michael Brantley batting against Jake Odorizzi.
0.284, 0.718 — Matt Kemp batting against Aaron Nola.
0.292, 0.715 — Andrelton Simmons batting against Ryan N Carpenter.
0.294, 0.714 — Ronald Torreyes batting against Andrew Cashner.
0.282, 0.712 — Eddie Rosario batting against Shane Bieber.
0.270, 0.710 — Albert Almora batting against Seth Lugo.
0.279, 0.710 — J.T. Realmuto batting against Jordan Lyles.
0.276, 0.710 — Nick Castellanos batting against Andrew Heaney.
0.278, 0.709 — Wilson Ramos batting against Daniel Mengden.
0.288, 0.705 — Yulieski Gurriel batting against Drew Pomeranz.
0.274, 0.704 — Starlin Castro batting against Jordan Lyles.
0.287, 0.702 — Manny Machado batting against Sonny Gray.
0.261, 0.702 — J.D. Martinez batting against Lance McCullers.
0.268, 0.701 — Starling Marte batting against Jack Flaherty.
0.265, 0.700 — Mookie Betts batting against Lance McCullers.
0.280, 0.698 — Adam Jones batting against Sonny Gray.
0.272, 0.696 — Matt M Duffy batting against Daniel Mengden.
0.274, 0.695 — Asdrubal Cabrera batting against Jose Quintana.
0.270, 0.694 — Odubel Herrera batting against Clayton Kershaw.
0.255, 0.694 — Jose Ramirez batting against Jake Odorizzi.
0.266, 0.693 — Corey Dickerson batting against Jack Flaherty.
0.242, 0.686 — Freddie Freeman batting against Tanner Roark.
0.256, 0.686 — Jose Martinez batting against Trevor Williams.
0.284, 0.684 — Miguel Andujar batting against Andrew Cashner.

Both systems are in agreement on Altuve and Segura as the top two players. Michael Brantley owns a 19-game hit streak, and has a .73 chance to extend it to 20 today!

Remember, your best pick will fail about 25% of the time. Good luck!



from baseballmusings.com https://ift.tt/2L9NH0d

Best Batter Today

Jose Ramirez reclaims the top slot in the Baseball Musings Batter Rankings after going two for four with a home run Wednesday afternoon. He slips past Mike Trout, who went one for three with a double and a walk in the Angels 6-1 loss to Detroit. Mookie Betts, J.D. Martinez, and Francisco Lindor round out the top five.

Over the last month or so, both Trout and Ramirez are getting on base extremely well. Trout leads in OBP, but Ramirez has more hits, and knocked out 25 extra-base hits. Ramirez also struck out just 14 times in 111 at bats. His putting the ball in play is a big reason for his success.

Mookie Betts is still recovering from his injury.



from baseballmusings.com https://ift.tt/2J0luMD

Thursday Update

The Day by Day Database is up to date.



from baseballmusings.com https://ift.tt/2kAtoxE

Wednesday, May 30, 2018

Games of the Day

Reynaldo Lopez leads the White Sox against the Indians and Corey Kluber. Lopez is keeping batters off base by limiting hits, but he’s not doing it with strikeouts. He’s avoiding good contact with only 18.3% of balls in play against him going for line drives. Kluber is only one game behind Max Scherzer for best winning percentage in the majors since the start of the 2016 season. They could be tied by the end of the day.

Dallas Keuchel faces Luis Severino as the Astros and Yankees finish their series. Keuchel, with a 3.39 ERA, has been reduced to the fourth starter on team as other Astros occupy the 1-2-3 slots on the AL ERA leader board. Severino improved on his excellent 2017 season by cutting his home run rate by more than half.

Rookies off to good career starts meet in Kansas City as Fernando Romero of the Twins battles Brad Keller of the Royals. Romero owns a 1.88 ERA after five starts. In 28 2/3 innings, he struck out 29 and allowed just one home run. That helps balance the 13 walks. Keller makes his first start after 21 relief appearances. He allowed 1 home run in 22 1/3 innings, walking just seven. He’s only struck out 13, however, but owns a 2.01 ERA.

Finally, Zach Eflin faces Ross Stripling as the Phillies play the Dodgers. Eflin takes a bend, don’t break approach with men on base. He walked just one batter with the bases empty, but six with runners on. His BA allowed, however, is lower with men on as he appears to pitch around problems. Stripling found a way to keep the ball in the park this season allowing just two home runs in 41 1/3 innings.

Enjoy!



from baseballmusings.com https://ift.tt/2J3qqvZ

Mets Bullpen Mess

The Mets are 2-7 over their last nine games, and a big culprit is the bullpen. They blew a 6-2 lead over the Braves Tuesday night as hit a walk-off home run in the ninth to win the game for Atlanta. The Mets starters are pitching well in that span, posting a 3.52 ERA with very good three-true outcome numbers. The bullpen is getting knocked around as if they were facing nothing but MVP candidates. They allowed a .338/.416/.547 slash line in 161 batters faced. Five of those losses were by one run, so even a moderately better result would have the Mets in the thick of the division race. Now they need to chase down three teams that are all playing very well.



from baseballmusings.com https://ift.tt/2JgubS5

Beat the Streak Picks

Here are the top picks my programs produced for use in Beat the Streak. This post mostly explains the ideas behind the calculations. In addition, this post shows tests on the Neural Network (NN). This post discusses an NN that includes the ballpark. I updated the models, and the results of those tests are here.

For 2018, I am just going to publish the Log5 hit averages and the NN probabilities with parks factored in. I am keeping track of the results here. I added a graph that gives a visual representation of the probability and success each day.

I have been asked to expand the list to the top 25 players for an econometric project.

First, the Log5 Method picks:

0.375 — Jean Segura batting against Matt Moore.
0.328 — Nick Markakis batting against Jason Vargas.
0.321 — Freddie Freeman batting against Jason Vargas.
0.320 — J.T. Realmuto batting against Clayton Richard.
0.319 — Andrelton Simmons batting against Michael Fiers.
0.314 — Starlin Castro batting against Clayton Richard.
0.313 — Ryon Healy batting against Matt Moore.
0.311 — Brandon Crawford batting against Jonathan Gray.
0.310 — Buster Posey batting against Jonathan Gray.
0.309 — Matt Kemp batting against Zach Eflin.
0.304 — Mitch Haniger batting against Matt Moore.
0.302 — Gorkys Hernandez batting against Jonathan Gray.
0.300 — Benjamin Gamel batting against Matt Moore.
0.298 — Ozzie Albies batting against Jason Vargas.
0.297 — Odubel Herrera batting against Ross Stripling.
0.296 — Brandon Belt batting against Jonathan Gray.
0.296 — Ender Inciarte batting against Jason Vargas.
0.294 — Evan Longoria batting against Jonathan Gray.
0.293 — Albert Almora batting against Joe Musgrove.
0.288 — Guillermo Heredia batting against Matt Moore.
0.288 — Brian Anderson batting against Clayton Richard.
0.286 — Kurt Suzuki batting against Jason Vargas.
0.284 — Miguel Rojas batting against Clayton Richard.
0.283 — Austin Jackson batting against Jonathan Gray.
0.281 — Nelson Cruz batting against Matt Moore.

Segura against Moore produces a huge probable hit average. Since the start of the 2016 season Segura hit .314 with a .358 OBP. Moore allowed a .272 BA and a .341 OBP. This year, however, Moore is at .344/.419 and that one year average looms large in the calculation. It is especially bad since the MLB hit average is very low, .222.

Here is how the NN with Park ranks the players:

0.375, 0.776 — Jean Segura batting against Matt Moore.
0.309, 0.733 — Matt Kemp batting against Zach Eflin.
0.320, 0.732 — J.T. Realmuto batting against Clayton Richard.
0.266, 0.731 — Jose Altuve batting against Luis Severino.
0.319, 0.730 — Andrelton Simmons batting against Michael Fiers.
0.328, 0.726 — Nick Markakis batting against Jason Vargas.
0.274, 0.726 — Scooter Gennett batting against Patrick Corbin.
0.314, 0.726 — Starlin Castro batting against Clayton Richard.
0.321, 0.726 — Freddie Freeman batting against Jason Vargas.
0.293, 0.724 — Albert Almora batting against Joe Musgrove.
0.263, 0.721 — Michael Brantley batting against Reynaldo Lopez.
0.310, 0.720 — Buster Posey batting against Jonathan Gray.
0.297, 0.716 — Odubel Herrera batting against Ross Stripling.
0.272, 0.711 — J.D. Martinez batting against Sam J Gaviglio.
0.276, 0.709 — Mookie Betts batting against Sam J Gaviglio.
0.275, 0.709 — Nolan Arenado batting against Derek Holland.
0.313, 0.708 — Ryon Healy batting against Matt Moore.
0.311, 0.708 — Brandon Crawford batting against Jonathan Gray.
0.270, 0.704 — Eddie Rosario batting against Brad Keller.
0.296, 0.703 — Ender Inciarte batting against Jason Vargas.
0.263, 0.701 — Charlie Blackmon batting against Derek Holland.
0.302, 0.700 — Gorkys Hernandez batting against Jonathan Gray.
0.259, 0.697 — Wilson Ramos batting against Sean Manaea.
0.265, 0.697 — Gerardo Parra batting against Derek Holland.
0.260, 0.696 — Starling Marte batting against Kyle Hendricks.
0.262, 0.696 — Asdrubal Cabrera batting against Julio Teheran.

The NN also produces a huge lead for Segura. The .776 is the highest probability for a batter calculated this season. Here are the parameters for the match-up, all hit averages (Hits/PA):

Parameters: [Pit 2018, Pit 2016-2018, Bat 2018, Bat 2016-2018, Non-pit LgAvg 2018, Park Avg. 2016-2018]
Parameters: [‘0.300’, ‘0.263’, ‘0.316’, ‘0.292’, ‘0.222’, ‘0.223’]

The three-year hit averages dominate the calculation, but when both one year numbers are high, as here, it gives the probability a big boost. Segura is just 2 for 10 against Moore in his career, but with only one strikeout.

J.T. Realmuto is the consensus second choice. Note also that Michael Brantley is working on an 18 game hit streak.

Remember, your best pick will fail about 25% of the time. Good luck!



from baseballmusings.com https://ift.tt/2LIPHh2

Best Batter Today

Mike Trout once again leads the Baseball Musings Batter Rankings, but Jose Ramirez has closed the gap between first and second place. Trout went one for four with a walk in the Angels 9-2 win over the Tigers, while Ramirez went two for four with a double and a home run in the Indians 7-3 win over the White Sox. About three points separate the players now. Mookie Betts, J.D. Martinez, and Francisco Lindor round out the top five. Betts has not players for a few days due to injury, and prolonged absences start to eat away at a player’s score. If he does not return to the lineup soon, expect him to start falling.



from baseballmusings.com https://ift.tt/2xoxyC3

Wednesday Update

The Day by Day Database is up to date.



from baseballmusings.com https://ift.tt/2L8OQVN

Beware Of “Experts” Sharing Data…..

I guess part of being perceived as being an expert is sharing data and research—all of which supports the conclusions you want people to reach.

Don’t get me wrong, we share data on the results we’ve produced, we share data on research we have conducted.  But we are careful to position the context of that data and how it might be interpreted.

But too often, we consciously or unconsciously, we “lie with statistics.”

I read an interesting article in Forbes.  Largely, I agree with the author, but he cited an example that is very misleading.  I’m certain there was no bad intent, but it was an example that reinforced his basic premise–everything in buying is going digital, the B2B buying process is becoming consumerized.

He posed an example, under the heading, “Don’t Get Fooled By The Occasional Analog Win.”

The example was behavior he noticed on a train ride.  A SaaS sales person, over the course of 4 hours and multiple conversations closed a single deal worth $10K.  He contrasted that with another sales person sending out a promotional tweet for a logo creation service to 1.2 M followers on Instagram and Twitter.  After blasting those messages out, at the end of the train ride, he had converted 800 to orders totalling $16,000.

Clearly, the conclusion we are supposed to reach is a simple posting on Instagram and Twitter was far more productive (in those 4 hours) than a series of phone discussions with a single customer.

“Experts” that want you to drink the “Social Selling Kool Aid” will revel in this example.  They will say, “Social selling produced 60% more revenue in the same time, QED!”

But one should be skeptical about these examples.  One should start questioning and looking deeper.  Some thoughts I had:

1.  The $10K SaaS deal was ARR.  In reality, the deal was far greater than that, an important metric in SaaS is CLV-Customer Lifetime Value.  In reality, this deal was probably worth $10’s of thousands in CLV.  I’m not sure what the entire sales cycle was, we’re left to think it was 4 hours, but even if that was the tail end of the sales cycle, that’s a great deal.  Let’s be really conservative and imagine CLV at $25K.

To create $1M CLV, I have to close 40 deals.  If it’s $1M ARR that I want to hit, I have to close 100 deals (but the CLV for that is $2.5M).  On a CLV basis, for the sales person, that’s a little more than 3 deals a month.  On an ARR basis, that’s a little more than 8 deals a month—both probably achievable.

2.  By contrast the Instagram/Twitter Deal was $16K, based on the description of the company, it looks like that is one-time revenue.  The results aren’t stunning.  Out of 1.2 M messages, 800 closed—or 0.067%, for an average transaction of $20 per customer.

One starts doing the math.  If I want to produce $1M in orders, I have to acquire 50,000 customers.  At the 0.067% conversion rate I must reach out to about 75M customers.  What’s this mean–can I go to the same 1M+ customers 75 times (clearly there will be significant fatigue in those customers receiving messages.  Do I cast a wider net, but the yields are likely to plummet.  How do I effectively reach those 75M customers?   And at what cost?

Now, all of a sudden, things look a little different.  Perhaps the “old school” method actually looks much more intriguing.

In fairness, it’s impossible to ascribe “goodness” to either case, though that’s clearly the implication the author is trying to make.  The situations are so different.  The selling/buying process, the target customers, the decision-making processes are very different.   For example the SaaS deal may have been a complex sale with multiple decision-makers and a higher level of risk.  The other deal is a very transactional deal, with very low risk, and likely to be a single decision-maker  (Brent and Nick, I know you are reading this, how many $20 deals require 6.8 😉 )

It’s impossible to tell which is better than the other, because these aren’t comparable situations.

Yet we are constantly presented with these false choices or alternatives.

And too often, we get sucked into believing them, often in very dangerous ways.

As consumers of this kind of data, it’s always important to be skeptical of the data and conclusions.  Make sure you understand the underlying context and assumptions.  Make sure those are relevant to you, representative of your business, customers and markets.  Be skeptical of claims of outrageous results, while they may be true, they may only apply to a very special case which may not be representative of the situations you face.

As experts, while it’s fun to present sexy compelling data, but if we are to be responsible and to engage followers in a meaningful way, we have to be careful how we position and present the data.  Thinking of customers a gullible and presenting data that, while true, may be very misrepresentative of what they might expect.  Be careful, it will always come back to bite you.

 

Afterword:  To help you understand and think about these issues both more deeply and broadly, I highly recommend reading the following:

Factfulness by Hans Rosling:  A stunning analysis of much of what we think of as true about world data on medicine, environment, economics, and how we look at problems in the world.  Lessons can easily be translated to what we do every day.

How To Lie With Statistics by Darrell Huff:  An absolute classic.  This was originally published in the 1940’s but is a fun read about how we use and misuse data and statistics.  Really a fun, fundamental book.

 



from Partners in EXCELLENCE Blog — Making A Difference https://ift.tt/2L7iwm2

Tuesday, May 29, 2018

Sales Manager, Are You Doing Your Job?

Monthly, I get into discussions with well intended sales managers–all are struggling, all are looking for help.  They are trying to do their jobs, but in most cases they don’t understand their jobs, so it’s no wonder they are struggling.

Conversations go something like the conversation I had several weeks ago with “Bob.”  Bob is a front line sales manager (FLSM) for a large technology company.  He has had a successful sales career and has been a FLSM for several years.

He’s making his numbers, but barely, he laugh’s, saying, “It’s always at 11:59 PM on the last day of the fiscal year.”

He works extraordinary hours, spending a lot of time on the road and on the phone.  The company, his management, also wants constant reporting and status updates.

As we started talking, I asked him, “Bob, what’s your job?”

Bob started looking at me, quizzically, I’m a consultant brought in by top management to help improve the organizational performance.  I know he’s wondering, “Shouldn’t he know this stuff?  Why’s he asking me such a basic question?”

Bob replied, “I’m accountable for making the numbers!”

I ask, “What’s that mean?”

Bob says, “I spend my time doing deals!  I’m constantly in the field and on the phone doing deals!  If I don’t do that, I won’t make the number!”

Bob talked about supporting his people, checking on what they were doing.  But everything came back to doing deals, most of the time with his people, but he was actively involved in the majority of the deals they were doing.  He felt if he wasn’t, he wouldn’t make his numbers.

Sound familiar?

Most conversations I have with sales managers, at least those that aren’t desk jockey’s hiding behind SFDC reports, sound exactly like this.

The problem is, they aren’t doing their jobs!  They are doing the job of their sales people!

The sales manager’s job is not “doing deals.”  There’s no doubt, they may need to be involved in many deals, supporting their sales people, or getting things done for their sales people.  But the job isn’t doing deals.

Let’s go back to that first question/answer, Question:  What’s Your Job?  Answer:  Making the numbers!  It’s not a great answer, but it’s the most common one.

The next question is, “How do you do that?  How do you invest your time in making sure you make the numbers?”

This is where the conversations start going off the rails, as with Bob, too many managers revert to their instincts as sales people–which is doing deals.  If they aren’t doing deals, they don’t think they will make the numbers.  But that’s really a strategy for failure–which is why so many managers are struggling.

In Sales Manager Survival Guide, I pose the scenario:  Let’s say that when Bob was a top sales person, he did 10 deals a quarter, always beating his quota.  He worked long hours to achieve that success.  Fast forward to Bob as a sales manager.  He probably has about 10 people reporting to him.  Each has to do 10 deals per quarter, or Bob’s team has to do 100 deals.

Bob isn’t involved in all the deals, but let’s say he’s deeply involved in 25% of them (usually, the Bob’s of the world feel compelled to have their fingers in pieces of all the deals.)  Now you can see how Bob’s model of doing things starts to break down.  Even if he is only doing 50% of the work on those 25 deals, it requires much more than what he did as an individual contributor—and he doesn’t have time to do anything else!

Now pile on the demands for forecasts, endless reporting, and other management requests.  You can see how Bob is drowning!

All because Bob doesn’t understand his job.

The only way a sales manager can “make the numbers,” is by maximizing the performance of each person on their team–enabling each person to make their numbers.  The sales manager can’t and shouldn’t be doing it themselves.

We don’t maximize the performance of our sales people by doing the job for them—“doing deals.”  We do this through:

  1. Making sure we have the “right people in the right jobs.”  Talent underlies everything.  If we don’t have the right people in the jobs, nothing we do will enable them to perform at the level we expect.  Developing rich competency models is critical to getting/developing the right talent.
  2. We have to make sure our people understand performance expectations and what “good” looks like.  This goes far beyond saying, “Here’s your quota, good luck and godspeed……”  It’s characterizing what people who are successful in your organization do–how they spend their time, who they spend it with, how you create value, how you differentiate your offerings/company, and so on.
  3. We have to make sure we provide our people the right processes, systems, tools, programs, and training to enable them to perform at the top level.
  4. And most importantly, it’s constantly coaching and developing them.  It’s helping them learn how to apply the processes, systems, tools, programs, and training in a way that drives success.  In this process, you will be involved in helping them do the deals–but not doing it for them.  Remember, “Feed a person a fish, they eat for a day.  Teach a person to fish, they are never hungry.”

In summary, the FLSM’s job is getting things done through their people!  Anything else cheats the people, and cheats the organization from achieving the full potential.

For any FLSM, there is an adrenalin rush in doing deals.  There’s ego gratification from doing deals.  If that’s your priority, go back to being a sales person.

If you are committed to being a top performer as a FLSM, make sure you know your job and execute it.

 



from Partners in EXCELLENCE Blog — Making A Difference https://ift.tt/2xxBwsb

The Year of the Trout

Read Devon Young writes that Mike Trout is listed at 4.9 rWAR at Baseball Reference. With the Angels now having played 54 games, 1/3 of the season, Trout is on an historic pace:

I know this is a big “if”, but If he stays healthy & keeps playing close to this pace, he could have a 12 rWAR season. According to b-Ref, there have only been five 12+ rWAR seasons by position players, & none in the past 50 years —
14.1 – Babe Ruth (1923)
12.9 – Babe Ruth (1921)
12.5 – Carl Yastrzemski (1967)
12.4 – Babe Ruth (1927)
12.1 – Rogers Hornsby (1924)
There have been 33 pitchers to post 12+ rWAR (1871-2017), but, most of that was from the 1800’s & early 1900’s, when it was a very different game. In the past 100 years (1918-current), there’ve been only 3 pitchers to pull it off —
12.2 – Dwight Gooden (1985)
12.1 – Steve Carlton (1972)
12.0 – Pete Alexander (1920)

Trout has a shot at surpassing Babe Ruth’s 1923 season. Note, too, that Mookie Betts is at 4.1 rWAR, so he has a shot at 12 rWAR as well.



from baseballmusings.com https://ift.tt/2kwbfkk

Castillo and EPO

Luis Torres at The Hardball Times explains everything there is to know about EPO, the substance that landed Wellington Castillo an 80 game suspension. This is the most telling paragraphs of the report, on the effectiveness of the hormone:

Let’s go back to the study examining the effects of EPO on cycling. Scientists designed a study that was as close to a clinical trial as they could. Ideally, you would want a few hundred experienced, professional cyclists. The problem is that there obviously is not a huge pool of these kind candidates to choose from, and of course, those that do qualify are prohibited from taking EPO at all. The best the scientists could do was “48 well-trained male amateur cyclists.”

The limitations of the study mean the results need to be taken with a grain of salt. That being said, the findings were striking. It appeared EPO was completely useless as a performance enhancer for cycling.

So if there is even the possibility EPO is ineffective for cyclists, how could it possibly help baseball players? And why would MLB ban it?

The ban has as much to do with the dangers of the drug as anything else. Castillo may have taken the drug for nothing more than tainting his career.



from baseballmusings.com https://ift.tt/2ITqUFa

Games of the Day

Charlie Morton goes for his eighth win without a loss as the Astros face CC Sabathia and the Yankees. Since joining the Astros in 2017, batters find Morton tough to hit. His .214 BA allowed ranks ninth lowest in the majors in that time frame. After allowing a 1.69 ERA in April, Sabathia gave up a 6.05 ERA so far in May. Despite a higher K rate in the month, Sabathia gave up more hits.

The Cubs try to push the Pirates further down in the NL Central as Jon Lester takes on rookie Nick Kingham. Lester owns an outstanding 2.37 ERA this season, but gave up four runs in five innings against the Pirates earlier this year. In his first 18 1/3 innings, Kinghmam walked just two batters while striking out 21.

Finally, Jake Arrieta leads the Phillies against the Dodgers and Kenta Maeda. Arrieta allowed just one home run at home and one on the road, but 29 K at home and just six on the road. The low K rate leads to a 5.12 ERA away. Maeda owns a 2.88 ERA at home this year versus 3.86 on the road. He’s allowed fewer hits and walks away in the same amount of innings, so the difference looks like good luck in Los Angeles.

Enjoy!



from baseballmusings.com https://ift.tt/2xnBRgF

Beat the Streak Picks

Here are the top picks my programs produced for use in Beat the Streak. This post mostly explains the ideas behind the calculations. In addition, this post shows tests on the Neural Network (NN). This post discusses an NN that includes the ballpark. I updated the models, and the results of those tests are here.

For 2018, I am just going to publish the Log5 hit averages and the NN probabilities with parks factored in. I am keeping track of the results here. I added a graph that gives a visual representation of the probability and success each day.

I have been asked to expand the list to the top 25 players for an econometric project.

First, the Log5 Method picks:

0.321 — Scooter Gennett batting against Zachary Godley.
0.315 — Mookie Betts batting against Marco Estrada.
0.315 — Jean Segura batting against Austin Bibens-Dirkx.
0.308 — J.D. Martinez batting against Marco Estrada.
0.303 — Eddie Rosario batting against Daniel Duffy.
0.302 — Wilson Ramos batting against Daniel Gossett.
0.300 — Jose Altuve batting against CC Sabathia.
0.299 — Asdrubal Cabrera batting against Anibal Sanchez.
0.295 — Nick Markakis batting against Steven Matz.
0.292 — Jose Martinez batting against Zach Davies.
0.290 — Matt M Duffy batting against Daniel Gossett.
0.290 — Freddie Freeman batting against Steven Matz.
0.290 — Odubel Herrera batting against Kenta Maeda.
0.279 — Nolan Arenado batting against Jeff Samardzija.
0.279 — Andrelton Simmons batting against Michael Fulmer.
0.278 — Marcell Ozuna batting against Zach Davies.
0.276 — Matt Kemp batting against Jake Arrieta.
0.276 — Mallex Smith batting against Daniel Gossett.
0.276 — Eduardo Nunez batting against Marco Estrada.
0.276 — Xander Bogaerts batting against Marco Estrada.
0.274 — Nick Castellanos batting against Nicholas Tropeano.
0.272 — Albert Almora batting against Nick Kingham.
0.270 — Manny Machado batting against Jeremy Hellickson.
0.270 — Mitch Moreland batting against Marco Estrada.
0.269 — Andrew Benintendi batting against Marco Estrada.
0.269 — Charlie Blackmon batting against Jeff Samardzija.

Gennett had another big day on Monday in a Cincinnati loss, going 3 for 5. Both Betts and Segura and injured but not on the disabled list, so watch to see if they are playing. Betts has left-side tightness, Segura a concussion.

Here is how the NN with Park ranks the players:

0.300, 0.747 — Jose Altuve batting against CC Sabathia.
0.315, 0.746 — Jean Segura batting against Austin Bibens-Dirkx.
0.321, 0.743 — Scooter Gennett batting against Zachary Godley.
0.315, 0.726 — Mookie Betts batting against Marco Estrada.
0.308, 0.725 — J.D. Martinez batting against Marco Estrada.
0.302, 0.724 — Wilson Ramos batting against Daniel Gossett.
0.299, 0.721 — Asdrubal Cabrera batting against Anibal Sanchez.
0.303, 0.720 — Eddie Rosario batting against Daniel Duffy.
0.268, 0.715 — Michael Brantley batting against Lucas Giolito.
0.295, 0.714 — Nick Markakis batting against Steven Matz.
0.290, 0.714 — Freddie Freeman batting against Steven Matz.
0.292, 0.713 — Jose Martinez batting against Zach Davies.
0.290, 0.711 — Odubel Herrera batting against Kenta Maeda.
0.272, 0.710 — Albert Almora batting against Nick Kingham.
0.279, 0.709 — Nolan Arenado batting against Jeff Samardzija.
0.279, 0.708 — Andrelton Simmons batting against Michael Fulmer.
0.274, 0.705 — Nick Castellanos batting against Nicholas Tropeano.
0.269, 0.705 — Charlie Blackmon batting against Jeff Samardzija.
0.276, 0.704 — Matt Kemp batting against Jake Arrieta.
0.290, 0.703 — Matt M Duffy batting against Daniel Gossett.
0.266, 0.700 — Jon Jay batting against Kyle Gibson.
0.266, 0.698 — Jose Abreu batting against Michael Clevinger.
0.278, 0.697 — Marcell Ozuna batting against Zach Davies.
0.268, 0.697 — Gerardo Parra batting against Jeff Samardzija.
0.276, 0.694 — Eduardo Nunez batting against Marco Estrada.

Altuve is two for six career against Sabathia in the regular season, two for nine if you include the post-season. Gennett is the consensus first pick.

Remember, your best pick will fail about 25% of the time. Good luck!



from baseballmusings.com https://ift.tt/2GZcq4F

Best Batter Today

Mike Trout hit his 18th home run of the season Monday to remain in top spot of the Baseball Musings Batter Rankings. Jose Altuve‘s time in the top five was short, as a 3 for 4 day by Francisco Lindor moves him back into the top five. Jose Ramirez, Mookie Betts, Lindor, and J.D. Martinez are two through five.



from baseballmusings.com https://ift.tt/2sm0KUh

Tuesday Update

The Day by Day Database is up to date.



from baseballmusings.com https://ift.tt/2IZFZ8h

Monday, May 28, 2018

Rookie Pause

Ronald Acuna goes on the disabled list:

Results from Ronald Acuña Jr.’s MRI on Monday showed a left knee mild ACL sprain and a left knee contusion as well as a lower back contusion. He will be placed on the 10-day disabled list and be re-evaluated.

Acuna injured his left knee as he completed his sprint to record an infield single during the seventh inning on Sunday. After crossing the bag, his left leg buckled in a gruesome manner and he immediately fell to the ground, where he remained while being attended to by members of the Braves’ medical staff.

Here’s hoping he recovers quickly.



from baseballmusings.com https://ift.tt/2JbN1df

Analyzing the Home Run Drop

Jim Albert at Exploring Baseball Data with R quantifies the drop in home runs this season. Using models based on the 2016 and 2017 season, there are many fewer home runs than expected based on launch angle and exit velocity of batted balls.

Note that the 2018 estimated probabilities are significantly lower than the 2017 values, even when you adjust for the early part of the season. In other words, the balls with a particular launch angle and launch speed are experiencing more drag in 2018, at least relative to 2017.

So it appears the drag problem with the ball was fixed.

Albert participated in the MLB sponsored study of baseballs. One of the recommendations of the study is this:

MLB should monitor and attempt to standardize the application of mud on the baseballs, since the surface texture of the baseballs affects drag.

The application of mud used to be the job of the umpires, but it has gone to clubhouse personnel over time. It’s labor intensive, as someone has to actually rub the mud on the balls. I suspect there should be a machine for this, some kind of tumbler like they use to coat candy. It might produce a more consistent ball.



from baseballmusings.com https://ift.tt/2IS5VXh

Thoughts On Agility

Agile is one of the “hot” business buzzwords these days, right along with digital, transformation, and disruptive.  (You get double buzzword bingo points if you can use, “disruptive, agile, digital business transformation” in a single sentence on a PowerPoint slide.)

But much of what I see under the “agile” banner is far from agile.

The concept of agility focuses on improving individual and organizational ability to respond to quickly changing business conditions.  Whether they are market shifts, competitive disruption, dealing with overwhelm and complexity (Internal, customer, market), being “agile” helps us both recognize these and take action to respond.  Alternatively, some approaches to agile would claim to drive those shifts/disruptions.

Velocity is an important concept in being agile.  It is the ability to recognize problems, challenges, understand and respond to them quickly.  But too often, we confuse speed and velocity.  Velocity measures the same thing as speed but has a directional component.  Executing with speed, being busy, being active without a direction is meaningless, wasted action.  It achieves nothing, in fact, it can take us further from our goals than doing nothing.

Yet, this is what we see, organizations executing lots of things, people who are crazy busy, but without direction or purpose.  This isn’t agile, it’s quite the opposite.

Empowerment and trust are key to being agile.  In order to respond to changes, quickly, we need to get people as close to the work/action as possible to recognize the need to change, and we must trust they will do the right thing in driving change activities.

Yet, “I have to check with my manager….” is the norm.  Or subtle “command and control” structures dominate.  No one likes that term, but it’s manifested in too often, in different ways:  “Do things this way, don’t change,”  “This is our process, you have to follow it,”  “Here is the script, make sure you stick to it,” “Check with me before you make a commitment,”  or “We need to go get approval for that….”

If everything has to funnel back to management, they become the bottleneck and the weak link in being agile.  There is no way management can possibly identify and respond to all that needs to be achieved in the agile organization.  There is no way that managers have the depth of knowledge/understanding on everything to be able to respond in meaningful ways.

Collaboration, fluidity are key concepts in being agile.  We “swarm” problems, we have “stand-ups,”  we keep everyone who needs to be involved and informed, aligning priorities and developing action plans to address.

Yet, too often, we see our organizations dominated by silos and departments.  These organizations are optimized around achieving individual, departmental, and functional goals–which may not be aligned with the expected end results.

Agility requires structure and process.  We cannot be agile in the absence of these. Structure/process provide context in which to understand which actions are most important and how to drive change.  At the same time, agility recognizes the need to continuously improve and adapt.

Without structure and process, chaos reigns, and agility is meaningless in a chaotic environment.

Agility requires purposefulness and direction.  Without these, we don’t know what we are trying to achieve.  At the same time, agility recognizes we have to adapt and change as the world around us changes.

Many think they are being agile by constantly shifting priorities, by adopting a “program du jour” mentality, changing aimlessly.

Agility requires alignment, across the organization.  If the entire organization isn’t aligned in goals, values, culture, priorities, it’s impossible to be agile.

It requires accountability, without accountability, it’s impossible to take meaningful action.

Agility only exists in an environment of constant learning and improvement.

Agility is about people–hiring the best, expecting the best, giving them the freedom to be the best and supporting them to achieve

There’s a lot more to being agile.  We aren’t agile just because we say we are.  We aren’t agile because we constantly shift goals, priorities, people.  We aren’t agile because we worship speed and quick changes.

There is a lot to being agile, unfortunately, too few understand and execute the things required to be truly agile.

 



from Partners in EXCELLENCE Blog — Making A Difference https://ift.tt/2IPMBd8