too small...
"A team source told SNY's Ralph Vacchiano that Oklahoma QB Kyler Murray is "probably a little too small" for the Giants.
The Giants "prefer to stick to the established measurables they have for a prototypical quarterback," Vacchiano notes. The Giants' organization emphasizes conventional wisdom and inside-the-box thinking, and they haven't started a quarterback that measures below 6-feet since 5-foot-11 Gary Wood went 0-6 in 1966. And "the philosophy hasn't changed all that much (in that over half-century timeframe)," Vacchiano confirms. 6-foot-3 Dwayne Haskins, 6-foot-4 Drew Lock, and 6-foot-5 Daniel Jones appear to be likelier candidates to succeed 6-foot-4 Eli Manning than 5-foot-9 Murray."
Source: SNY.com
I'm not dismissing "analytics", just questioning the constant barrage of accolades for anything the Pats do and how easy it is to duplicate. Its pretty funny.
I'm not dismissing "analytics", just questioning the constant barrage of accolades for anything the Pats do and how easy it is to duplicate. Its pretty funny.
There is a ton that can be learned from the Pats. Saying otherwise is like saying you couldn't run the West Coast Offense without Joe Montana and Bill Walsh.
If you want a good (though depressing for Giants fans) read, check out Mike Lombardi's "Gridiron Genius". Bill Belichick isn't doing this through wizardry and dark spells. There are aspects of what he does that can be studied and learned from. Does that mean it's possible to recreate the Pats' success? Probably not. But for a 5-11 team with miles of room for improvement there is a lot that could be applied and clearly is not.
Yes but this is the problem with this thinking and what the Giants actually did. Teams like the Pats and Eagles have had this programs for a decade plus. Many others for 5+ years. When the Giants conducted their GM search did they even bring in someone from the Pats? Eagles? From the Jags?
No. They went in the opposite direction and hired someone that produces headlines that make us look like we are living in the past and didn't even really do great diligence or information collection on other avenues.
You know what frequently is the enemy of progress? The difficulty of a problem. But you eat an elephant one bite at a time.
You can bring in someone that you've seen eat chickens before, and sure it shows that they can eat things. But no matter how many chickens they eat the elephant isn't going away.
One of my favorite stories from my time in AI was a prospective client, very large data provider. We did an MVP with them and I pitched to a room of 20 people. We told them our price for a go forward and we were told while they were impressed with our capabilities there was a big enough contingent in that room that "didn't know if they wanted to be in the analytics business." 3 years later they paid 500X what we were going to charge them to acquire a company.
Heads rolled for that. The people on the wrong side of that debate lambasted. The guy on this thread that said you have to invest in this for 5 years wasn't wrong. It might not pay dividends today or tomorrow but you have to pay the technology piper. It is coming, it's already here for many teams and there is ample evidence we are among the laggards here. Zebra data linked below, is not being leveraged for "data analysis" that Ty Siam is doing. Make no mistake about it this is needed for complex predictive modeling and game theory simulations by many teams.
It appears instead of trying to hasten our efforts to catch up we have people in and out of the organization minimizing this problem. I'm not even saying Gettleman isn't capable of bringing this team back to the playoffs in the next few years (although i'd place a higher probability on teams that have these analytics programs) but what I am saying is unless we change our course and bring in real technology leaders our performance over the next 10-15 years is going to be disappointing relative to the rest of the league.
Zebra Data - ( New Window )
It's beyond bizarre at this point.
This fucking guy literally determines what a person's competency is based on a LinkedIn profile!
Luddites might not place much stock in google searches or LinkedIn profiles but people that like to analyze information and build as complete of a picture about the job they want to take do. People that build advanced models like to collect whatever information is out there and what is out there does not put the Giants in a positive light.
Some of us like to back our points up with research and subject matter knowledge. Others of us like to make the same points over and over again, argue semantics and put our heads so far up GMs asses that it's hard to tell if they have original thoughts of their own.
You can use LinkedIn to get a sense of a job you want to take, then you might meet the person you researched and find it is absolutely not a person you want to work for or a company you don't want to join.
You still don't get it. You have based sweeping comments on the competency of people based on LinkedIn? You don't see that as being fucked up? I'm assuming you don't, since you not only keep posting the profiles, but use it as evidence of who is qualified and who isn't.
Luddites also never explain why when you google other teams vs. the Giants there are much more positive things written about their analytics programs.
What about what I posted above when I googled the Seahawks analytics program the article went out of it's way to use the Giants as a bad example of a team embracing analytics.
Like Luddites always do they zero in on one specific part of an argument while failing to acknowledge that not only is that single point a huge nit but the rest of the argument still holds strong.
If you'll recall, Shurmur went for 2 against Atlanta in a spot where conventional wisdom says not to.
That was based on data and analytics. That data told him going for two there would increase the teams chances of winning by 15%. Something like 60% if the defense had been able to actually get the ball back - which they struggled with all year.
Again, he implies that he has the type of data you seem to be lamenting NYG being resistant towards here...
SI put out this article in 2017 regarding how all 32 teams handle analytics. Here's what they said about NYG...
It's strange, because it seems like what I am reading isn't jibing with what you're saying. You're kind of giving off the impression that NYG have Fred and Barney hammering out numbers on a giant boulder while all 31 other teams are working with the latest, cutting-edge tech. That's not the impression I'm getting.
It seems to me that way too much was made of a Dave Gettleman quote and that became a narrative that isn't even really true.
The Giants have had analytics people in place since before Gettleman was even here. Do you think he sent them all home and told them he doesn't believe in that junk?
Well, if you read what Shurmur said, that's clearly not the case.
Yes but it really doesn't do you much good, it's like lying on your resume.
Also there are services you can buy, yes Shurmur is using some things but it is really just a slice of what is available. It's like someone told him there is something out there that tells you whether to go for 2 or not and he is using that.
Timeouts are actually more simple than two point conversions from a game theory perspective yet Shurmur did seem to have the resources in place to get that right.
And also yes if you look at Ty Siam's qualifications vs. Sean Harrington's they are night and day. One has prior experience using tools like STATA and Excel and the other has a computer science degree, makes original autoencoding algorithms and has engineering experience at Raytheon and IBM.
If I were an expert in analytics, I'd put the probability that somebody who draws sweeping conclusions from Google and repeatedly uses the term Luddite and "smelling farts" as being competent at 0%
Captain Analytics!
Quote:
but couldn't an incompetent person hire a competent person to do an amazing LinkedIn profile for them?
Yes but it really doesn't do you much good, it's like lying on your resume.
You'd still have to waste your time vetting out all these so-called competent folks your analytics claimed were great prospects. (I'm not against analytics, they can be very useful.)
Even a 3rd grade teacher would laugh at that!
It's fun! Luddite Luddite Luddite!
If you have your choice of team to interview for and work for you will want a team with evidence of buy in.
I'm not saying this is the only basis for your determination but also if there is smoke there is fire. I'm sure interviewing with Ernie Adams vs. Gettleman in terms of your belief in the resources you get and how the information is going to be used is way more of a positive experience in the Patriots building.
No this isn't the only information available but it is more likely that you will A) want to take the interview with a team like the Pats and B) actually have a good experience in that interview that makes you want to take the job
Even a 3rd grade teacher would laugh at that!
Google searches are biased to some degree that will ultimately benefit Google's bottom line.
Are you sure you aren't trolling at this point?
You really are asserting that Google searches are giving you all the insight you need, and you're a supposed expert in the field?
The fuck you are
Also there are services you can buy, yes Shurmur is using some things but it is really just a slice of what is available. It's like someone told him there is something out there that tells you whether to go for 2 or not and he is using that.
Timeouts are actually more simple than two point conversions from a game theory perspective yet Shurmur did seem to have the resources in place to get that right.
And also yes if you look at Ty Siam's qualifications vs. Sean Harrington's they are night and day. One has prior experience using tools like STATA and Excel and the other has a computer science degree, makes original autoencoding algorithms and has engineering experience at Raytheon and IBM.
What are you basing your "perception around the league" stuff on? Google searches?
I'm guessing you're not actually in contact with people in the league, so... that's a shoddy claim at best.
Based on my limited research, I think I made a reasonably compelling case that while the Giants are certainly doing some things with analytics they are not applying broadly. They use it from managing player health things like hydration, electrolytes and measuring various other physical markers to try to improve peak performance. They are also clearly trying to use it for things like when to go for 2 points...
That the Giants are do not seem to be using it for, nor are they appropriately staffed to attempt it is to project positional value, and map past player performance to potential performance with the Giants. For example, the Patriots are way ahead in this area.
And that's just the point the great job they think he's doing to someone with real knowledge of technology could be thoroughly mediocre or even bad.
Beyond that, the results on the field do not speak to anyone doing a good job. This is the problem with what you continue to argue, if the Giants were doing well, maybe you'd have a point, but they are decidedly not doing well. Not only overall but again, when they are fucking up simple game theory applied to taking timeouts in game when this guy has been employed for 3 years it definitely suggests that he doesn't know how to apply simple game theory to building models, so what exact reason do we have to believe he is doing even a mediocre job?
The results on the field and the skills in his LinkedIn profile match the thesis that the Giants don't have the type of people in the building to implement software and math in a beneficial way. I am bringing this evidence to the table again, what is your evidence to the contrary?
Other than trying to attack my methods for information gathering or my qualifications which again fall short in their actual evidence vs. fruitless banter and bravado.
"Limited research," "seemingly," "looks like"...
These are all modifiers you use when you want to make a point but don't quite have the necessary information to prove it.
I need to see more than LinkedIn profile links before I mindlessly just believe that the Giants are rejecting analytics and don't want it to be a part of their way of doing things. That's basically the impression people are trying to give off here.
If you think they're not as far along as other teams - that's a different argument and hardly seems to warrant the diatribes and crusades here.
Quote:
searches that it's not just what Gettleman said, there is perception around the league that the Giants are behind.
Also there are services you can buy, yes Shurmur is using some things but it is really just a slice of what is available. It's like someone told him there is something out there that tells you whether to go for 2 or not and he is using that.
Timeouts are actually more simple than two point conversions from a game theory perspective yet Shurmur did seem to have the resources in place to get that right.
And also yes if you look at Ty Siam's qualifications vs. Sean Harrington's they are night and day. One has prior experience using tools like STATA and Excel and the other has a computer science degree, makes original autoencoding algorithms and has engineering experience at Raytheon and IBM.
What are you basing your "perception around the league" stuff on? Google searches?
I'm guessing you're not actually in contact with people in the league, so... that's a shoddy claim at best.
Arc, I've posted the articles and summaries but the fact remains if you google New York Giants analytics vs. many other teams there are a lot more positive things written about other teams. I do think if the Giants were doing a good job in this department the news coverage would be much more positive as it is with those other teams.
And let me just reiterate - my sole reason for posting on that thread was to refute the idea that Gettleman shunned analytics and DID NOTHING in Carolina.
I've never made the claim that the Giants were at the forefront of the league or that they can't improve - it was simply to say they are doing something - a point Captain analytics was refuting and then shifted the argument to LinkedIn qualifications to state that we had the wrong people in place
A lot of these points are reaches...
And enough with the fucking luddites stuff, dude - it's making me not want to even continue this conversation because it's that obnoxious. We get it...
Quote:
In comment 14296883 NoGainDayne said:
Quote:
searches that it's not just what Gettleman said, there is perception around the league that the Giants are behind.
Also there are services you can buy, yes Shurmur is using some things but it is really just a slice of what is available. It's like someone told him there is something out there that tells you whether to go for 2 or not and he is using that.
Timeouts are actually more simple than two point conversions from a game theory perspective yet Shurmur did seem to have the resources in place to get that right.
And also yes if you look at Ty Siam's qualifications vs. Sean Harrington's they are night and day. One has prior experience using tools like STATA and Excel and the other has a computer science degree, makes original autoencoding algorithms and has engineering experience at Raytheon and IBM.
What are you basing your "perception around the league" stuff on? Google searches?
I'm guessing you're not actually in contact with people in the league, so... that's a shoddy claim at best.
Arc, I've posted the articles and summaries but the fact remains if you google New York Giants analytics vs. many other teams there are a lot more positive things written about other teams. I do think if the Giants were doing a good job in this department the news coverage would be much more positive as it is with those other teams.
This isn't evidence, I'm sorry.
A lot more positive things written about other teams than the Giants is your measuring stick?
I'll need better than that.
They've won 8 games in the last 2 years.
It's nothing more than that.
"Limited research," "seemingly," "looks like"...
These are all modifiers you use when you want to make a point but don't quite have the necessary information to prove it.
I need to see more than LinkedIn profile links before I mindlessly just believe that the Giants are rejecting analytics and don't want it to be a part of their way of doing things. That's basically the impression people are trying to give off here.
If you think they're not as far along as other teams - that's a different argument and hardly seems to warrant the diatribes and crusades here.
I use those limiters... Yes...
Perhaps the Giants have a super secret analytics department somewhere that nobody knows about. Highly unlikely but possible. So yes I use such verbiage, but don't mistake it for me making a weak point. In the thread I reference, I used such verbiage as well, but I was clear then that while the possibilities exist, I gave significant evidence that it was highly unlikely that the conclusion drawn were significantly off the mark.
The LinkedIn qualifications, Gettleman's comments, the more positive articles written about other teams they all coalesce to a story that the Giants are behind.
There are many teams that tell a more positive story and their executives outright praising analytics and their role in success.
I think it is also naive to expect this to change without the Giants adding new technical leadership. What exactly is your problem with saying you know what it would be a super positive thing if we made a high profile engineering hire and gave them resources?
They've won 8 games in the last 2 years.
It's nothing more than that.
There was plenty of press about Cleveland when they were worse than the Giants...
There is plenty of press about the Easgles before they became SB champions.
When I was looking I easily found press about Huston, Baltimore, Pittsburgh, and KC. THere were others as well, I just can't remember them all...
And the press goes back several years not just 1 or 2. The information about the Patriots goes all the way back to 1999!!!
Ask Sean Harrington, I am sure he has boiled it down to to a normal curve projecting his likelihood of success... But its all percentages.
I've said it on this thread and I'll say it again. My original point was the fact that they couldn't grasp the game theory in proper timeout usage did not reflect that they were leveraging analytics at all. (especially this information used in conjunction with what Gettleman said)
Now certain Luddites (and i'm talking about a fat Luddite and a British Luddite) both called me stupid for making this point. I reacted with aggression in kind and that's how we got where we are.
The fact remains that some luddites want to pretend like the way the Giants are approaching analytics isn't problematic and then there are some of us more informed in the understanding of the benefit of analytics and what is coming through the pipe data and technology wise that are in strong contention with that assertion. And those people might have taken issue with being called "stupid" when bringing it up.
This may be the very first time I find myself making any point counter to something he is is saying.
arc in general reflects my views almost exactly.
While he is still FMiC, I have noticed that he has become far less nasty over the past 8 months or so. Back then most posts he made was derisive towards somebody, and he was not provide much in the way of analysis.
His recent posts are much more interesting, thought provoking and based on a reasonable factual basis. His general position is far more team and especially front office friendly than my opinions though. The change probably has nothing to do with our interaction last spring, though I would like to think so ;).
It's not an easy thing for a lot of the old school guys to grasp. Even Sean Payton says he wishes he'd go with his own intuition more often rather than default to the data because he still doesn't totally trust it even if it's generally embraced it.
My impression is that NYG have indeed embraced analytics, but perhaps their department isn't quite as deep or extensive as others. Perhaps we're playing catch up. I wouldn't write any of that stuff off - it's just something that's difficult to quantify.
Even SI alluded to NYG being "quiet" about it despite being more "aggressive" behind the scenes back in 2017.
I would guess that certain teams have embraced this earlier than others - like NE - and most of the league is starting to play catch-up now.
Akin to MLB... I know the Yanks, Red Sox, and Astros in particular all place tons of emphasis on analytics and forward thinking data and those teams are ahead of a lot of others right now.
I've said earlier that I believe this is much more vital in baseball, but obviously I am in favor of the Giants using any data available to them to enhance their operation and increase their odds of success.
So, I don't necessarily think NYG are against employing analytics - obviously they aren't. Pat Shurmur said several times that they are using them. Are they behind the teams that are succeeding more than they are? That's certainly possible. I'd buy that.
But then, at that point, why is the ire directed at Gettleman?
It's not like he got here and scaled the analytics operations back or shut down the department. We seem to have just not been one of the teams at the forefront - which is true of the majority of the league.
Perhaps there are areas we could invest more in.
Bottom line - analytics are still relatively new in the NFL and a lot of people are still working with the data they're collecting to determine the most effective ways of applying it, using it as predictors, etc. - it's sort of a dangerous game to use the Patriots as examples for things because of how anomalous they really are.
I'd certainly hope that our people were constantly looking at what they do and looking to employ similarly successful strategies - but I think by now, it's safe to say most teams have probably done that or been doing it and there's just no way to replicate Bill and Tom.
It's not an easy thing for a lot of the old school guys to grasp. Even Sean Payton says he wishes he'd go with his own intuition more often rather than default to the data because he still doesn't totally trust it even if it's generally embraced it.
My impression is that NYG have indeed embraced analytics, but perhaps their department isn't quite as deep or extensive as others. Perhaps we're playing catch up. I wouldn't write any of that stuff off - it's just something that's difficult to quantify.
Even SI alluded to NYG being "quiet" about it despite being more "aggressive" behind the scenes back in 2017.
I would guess that certain teams have embraced this earlier than others - like NE - and most of the league is starting to play catch-up now.
Akin to MLB... I know the Yanks, Red Sox, and Astros in particular all place tons of emphasis on analytics and forward thinking data and those teams are ahead of a lot of others right now.
I've said earlier that I believe this is much more vital in baseball, but obviously I am in favor of the Giants using any data available to them to enhance their operation and increase their odds of success.
So, I don't necessarily think NYG are against employing analytics - obviously they aren't. Pat Shurmur said several times that they are using them. Are they behind the teams that are succeeding more than they are? That's certainly possible. I'd buy that.
But then, at that point, why is the ire directed at Gettleman?
It's not like he got here and scaled the analytics operations back or shut down the department. We seem to have just not been one of the teams at the forefront - which is true of the majority of the league.
Perhaps there are areas we could invest more in.
Bottom line - analytics are still relatively new in the NFL and a lot of people are still working with the data they're collecting to determine the most effective ways of applying it, using it as predictors, etc. - it's sort of a dangerous game to use the Patriots as examples for things because of how anomalous they really are.
I'd certainly hope that our people were constantly looking at what they do and looking to employ similarly successful strategies - but I think by now, it's safe to say most teams have probably done that or been doing it and there's just no way to replicate Bill and Tom.
I think the ire towards DG is based on his comments at the presser for the Barkley choice...
While it is clear that the Giants do use analytics, there is no doubt about that, it appears that they reject it as an aid for team building. Using as an aid for team building would be extremely complicated and involve an incredible number of variables. TO do it effectively, you need a staff of people, 20 on the low end, more likely about 30 - 40. There are teams where you can verify that they have such staffs. You can also find out the department leads for the staffs. They often involve people with advanced degrees in computer science, big data analysis, statistics, machine learning, and wall street type trading analysis.
That's the point about Ty Siam, he is the face of NYG analytics and his resume is weak at best. There is no technical staff to do software development, operate a data center, manage databases, or analyze the data. The Giants are not staffed appropriately to even attempt this sort of thing. So it appears they have no intention at this point in moving in that direction.
Many of us find this to be frustrating because we feel the team will make better personnel decisions if they did.
I go much farther than even that. VR technology should be used to get simulated reps. Hell the president installed such a system in the WH so he can play golf in the residence.... Why can't the Giants use something similar but designed for football!
Analysis can be done about play effectiveness, Sequencing plays, how best to isolate your most favorable matchups. I can go on forever in the ways in which technology can help the team.
But all of this requires serious technology leaders and staff. None of which the Giants have.
While he is still FMiC, I have noticed that he has become far less nasty over the past 8 months or so. Back then most posts he made was derisive towards somebody, and he was not provide much in the way of analysis.
His recent posts are much more interesting, thought provoking and based on a reasonable factual basis. His general position is far more team and especially front office friendly than my opinions though. The change probably has nothing to do with our interaction last spring, though I would like to think so ;).
Yeah perhaps if I was able to take a more measured approach originally I could have contributed to those efforts but I tend to respond equally disrespectfully when i'm called stupid and talked down to on a subject matter I've devoted a good deal of time to understanding and applying. The high road always works better though, you have done a better job of that in these debates than I. Honestly, there is an element to it where doing AI as a service for 6 years before my new venture I get to say things here in the confines of the internet that I can't say there.
Either way I've been on the board for a long time and don't have a lot of patience for bully's in general so there were multiple factors at play.
I do think it is good to keep in mind that we all have the same goal here the toughest thing for me to comprehend is why there is so much desire to trust the team and take them at face value despite recent difficulties and decidedly bad PR as it relates to analytics.
Rivera talked about 2015 being the first year Carolina really started to employ analytics - Gettleman was in Carolina then. So, this is his 2nd consecutive stop where analytics are being employed.
I have no idea what his plans are going forward, but I would assume the intention is to continue to scale it and not just leave it stagnant.
It's like Gruden - he ripped analytics earlier in the year, and then quoted PFF data later on - proving that they are indeed looking at it and utilizing it. These old school guys just don't like change. It's like Clint Hurdle in Pittsburgh - Huntington told him they were going to move in a direction that embraced analytics and that he was only going to be able to stay on board if he was willing to adapt and change.
He was resistant to it at first, but eventually learned to understand the value and work with the front office in terms of employing more new-school thinking.
It's actually why they went after Russell Martin years ago - data they had showed that he was one of the best pitch framers in baseball and got a higher percentage of borderline pitches/balls called strikes and that the increase of strike calls had a specific long-term value making him a worthwhile investment.
I'm getting off track here - but you get the idea.
It's sort of a scramble right now. I think teams are collecting data faster than they can figure out how to even effectively employ all of it - this is all happening on the fly.
Regardless - I'm not against anything that involves expanding the way the Giants view things. I just don't necessarily think they're as resistant to moving in this direction as perhaps they're being made out to be here.
If this is an arms race is isn't very smart to be on record insulting the very people you need to bring into the fold for said arms race.
It isn't just that Gettleman said that either, he really doesn't seem like he has the personality to make people feel empowered in the way that technical people like to feel. The best engineers I know tend to think differently than "business" people and while I think those strengths and weaknesses go hand and hand it is important first and foremost that the more complex needs and different work styles are paid with respect. I think this article on "straders" or traders who code outlines it well. There are reasons that there are areas of GS now that don't have a dress code or strict work hours. Good code is written with patience, an open mind and an outlook that doesn't look at things in a binary way. This goes doubly for applying the information generated from that code.
Gettleman very much seems like a my way or the highway kind of guy who makes up his own mind and does not empower the people around him. The public information that we've outlined on this would suggest that other teams are much friendlier to analytics and a culture of innovation and there isn't really any evidence to the contrary provided by anyone about the team or Gettleman.
While i'll admit that this doesn't constitute proof it certainly points to the Giants having more of a problem than a solution.
At one point, I thought there was an agreement that NOBODY knows for sure how the Giants are equipped to use analytics. Now we've reverted back to all the "evidence" being provided through exhaustive Google searches and LinkedIn profiles.
I think this goes again to a different way that technical minds think. Part of the open source movement is collaboration but the other part is to see what people can build with your code in order to recruit them or expand their ideas. Google makes tensor flow available not just out of the goodness of their heart and a belief in community.
The fact that the Giants might have a "secret operation" actually speaks more to a backwards way of thinking than something that should be looked at positively.
You don't want to give away the farm and google is never ever going to put their best stuff out there but the fact remains if you want to recruit you need something to draw people in show you have something exciting to work on in addition to the right culture.
The bottom line is if there is any team that should be most concerned about secrecy it's the Patriots as they appear to be ahead of the pack not only in what is written about but the results on the field.
Everyone talks about how BB is the best but did anyone stop to think that his partnership with Ernie Adams may be inextricably intertwined with that?