Jump to content

Noozak

Burnt Ends
  • Posts

    151
  • Joined

  • Last visited

Posts posted by Noozak

  1. 15 minutes ago, closetojumping said:

    Knowing Connelly, it's probably designed in part to protect him from SP+ picks that blow-up on any given Saturday. The guy is deeply insecure when it comes to questions around his work, and him creating new "data" that reinforce why his initial model works is pretty much right up his alley. 

    "What happens if the housing market goes down in value for a quarter?"

    "It won't."

    "So that's not in your quant's model?"

    "No. The model only factors in facts around the housing market."

    "But the housing market went down in value multiple times in the past."

    "Not since Hitler invaded Poland. Can't happen."

    Whoops. 

    "What happens if number of plays actually called has a disparity of greater than 25?"

    "That won't happen."

    Whoops. 

    The bottom line is that the disparity in plays called in this game was a direct indicator of the futility of ATM's offense and defense, in far greater ways than YPP was. It indicates that they were dominated on both sides of the ball. Any idiot can assemble logic in their model to discount or negate something like total plays until the disparity exceeds X, and then that gets factored to greater and greater levels as X increases. There is a lot of shit that doesn't matter in an average game that can be defining in major upsets or blowouts. 

    Specifically, think about it in this context: Team A runs 65 plays. Team B runs 50 plays. Which team probably won? The answer is: Tell me each team's YPP. If team B averaged 12 YPP, I'm guessing they might have won.

    Now try this one: Team A runs 80 plays. Team B runs 40 plays. Which team probably won? Anyone answering "Team A" has plenty of logic and supporting history to justify the choice, YPP unseen. There are undoubtedly games where someone averaged 15 YPP and only ran 40 plays, but way more often than not, they ran 40 plays because they were being dominated on offense. I'm being lazy, but what was the play gap between Texas and KU when we beat them 59-0? ATM vs. OU in the 77-0 rout? I'm guessing that those are also wide fucking gaps because or pure LOS domination. I don't know, maybe I'm off my rocker. 

    ATM was dominated upfront from whistle to whistle. They're lucky they didn't lose by double digits. If a model can't take that into account and attempts tell us that there is only a 22% win "probability" for App State in a game like that, then the model is shit and worthy of nothing but open derision.

    I agree with you. I think it really all boils down to your first point about his insecurities. Every model has flaws because it is impossible to build the perfect model. If I could build the perfect model I would be worth more than Elon Musk. Shit, whenever my data scientists go through the results of their models, I have them start with why their model sucks. As long as it doesn't suck too bad then the results typically don't matter. It provides some insight that may move the needle a little bit and we go on justifying our salaries. 

    On the other hand though I will give him a small bit of leeway (5% or so) though based on how the general public thinks about probabilities. 

     

  2. 40 minutes ago, closetojumping said:

    I don't want to put too much on the aggie schedule outcomes over the next 4 weeks, but it is hard not to do so. If ATM shits the bed and goes 1-3 or 0-4 during that run, there is going to be a lot crawfishing coming from the people participating in buying players. They're trying to fuck with Texas on Muhammed. With MM, they're offering something absurd, but it's not clear that they can even get that done. Texas wants MM no matter what, in any event. 

    If ATM is in Year 5 of Ritz Bit and they are sitting at 2-4 and trying to push for recruits, they're not going to have as much firepower behind them. To say that the money this cycle is being reluctant would be an understatement. 

    Also, the word is that they're more interested in seeing Hale not go to Texas than they are in having him sign with them. I mean, in any other setting, I would say that sounds like pure bullshit. Of course, this is not any other setting.

    Wait, so you think Bama looking like shit at WR is a good thing in recruiting Hale? I don't see that. Sarkisian and Marion have everything they need in order to sell Texas to recruits, Hale included. 

    Do you think someone like Anthony Hill is shut down regardless of the season aTm has? 

  3. 18 minutes ago, bigup2dahorns said:


    Baxster is from Orlando FL and goes to Edgewater HS in Orlando, not in Edgewater FL. Not sure where Carson is now and if anyone can still pick him up.

    Yeah Edgewater is located in College Park a suburb of Orlando. I moved to Orlando mid way through high school and their fans were known for throwing rocks at opposing teams. We did beat them to go to the state semi finals my junior year and we attempted to steal their goal post before the police shut it down.

    • Hook 'Em 3
  4. That experience was Peak Miami. Someone was playing a set in between every major intermission. There was always something to do. There were tons of food local food stalls that were better than expected. God it was so fucking hot though. Thankfully they had water stations everywhere. Like someone else said, ride share was fucked and the lines were insane. On Sunday we paid someone to park in their yard north of the stadium. It cost me $50 compared to he $300 on stubhub or ticketmaster. We got out after the race and after a 20 minute walk we got right on the highway with almost no traffic. I heard from tv watchers the race was boring but I sat at turn 1 and it felt like every lap there was something going on. I ended up paying $3000 for two, 3-day passes at turn 1 and I didn't regret it one bit. 

    I did hear the people who paid for paddock passes were disappointed in the food quality, and that the cabanas were a giant waste of money so definitely a mixed bag in some places
     

     

    A98C468E-41CE-41F0-9904-5809BC797206.jpeg

    • Hook 'Em 1
    • Like 1
  5. Anyone know if you can resell a single day from a 3-day ticket pack? Looking into some last minute tickets to Miami and if I can make some money back then It would be worth it. 

  6. 7 hours ago, Homercles said:

    Holy shit I wished I’d have seen this thread three years ago when I’d just finished the remote Masters program there…it was called Predictive Analytics then but probably renamed by now as that term seems to have moved onto the next buzzword like data science or machine learning.  
     

    What stage are you in?  I graduated in 2017, and the program was very R-heavy which I regret as Python has taken over as a much more user-friendly, general programming language over the sometimes-nasty R which behaves much more like a calculator scripting language and is hard to integrate or scale.  
     

    The hardest and most rewarding course I took was with Dr Bhatti, maybe it was 422.  we branched into much more applicable topics such as boosted trees and and legit machine learning approaches like neural nets.  He was my favorite professor by far…tough but fair, engaged and expected you to learn the material ahead of time like a true graduate student.  
     

    The program overall came at a bad time as it was already behind the progression towards cloud-based, production-focused, scalable techniques that let you crunch much more data than that baby csv file of Iowa housing prices they we explored early in the program.
     

    Sadly I never really got to apply much of my learnings in data science in my role as I’m much more of a program manager and executive communicator than legit data scientist. I’ve forgotten most of the theory and details, but my main takeaways working on the fringes of data science are:

    1) Most people don’t spend nearly enough time applying the scientific method to really define the problem, break it down into testable hypotheses and applying the appropriate technique given the data available.   They want answers…now…not to really define the problem.

    2)  There rarely is a magic answer that’ll solve all your ills in business.  You have to be multi-disciplinary in fully understanding how the data you are using is generated, what it reflects, how data science applies their workflows to it, framing results in those contexts and applying decision science with the limitations in mind to move forward.  
     

    3) My jobs main focus is now in data engineering, as it’s the foundation of any applicable uses of the data…reporting, baseline analytics, predictions, production applications and effective usage of it.  We grew our data scale too quickly and a mix of legacy systems, lack of comprehensive data quality programs, automation and wild mix of volume, velocity and veracity make for a lot of cleaning up ETL processes.  
     

    4) Therefore my team is implementing ‘Data Ops’ this fiscal year as a culture, tools and workflow design ethos to treat data engineering like a true factory process…lean dev methodologies, Six Sigma continuous improvement and tightly integrated production stacks.  Azure is our preferred platform as we are an MS shop so DevOps, Azure AD, Data Factory / Azure SQL / Databricks, Power Suite are all tightly integrated.  Google Data Ops for more info.  
     

    5) Leadership needs patience.  For every successful implementation of a model there are a dozen that didn’t prove fruitful.  Nothing works like the movies.  Your data scientists need clean, raw data with knowledge of the business for feature engineering whereas analysts/reporting need simple structured summarized data to be effective at a lower skill level.  
     

    Sorry for the novel.  I hope the NU program has advanced since I went through it because the world has changed rapidly, using R on a 100k-row CSV file won’t cut it anymore.  

    Yeah they renamed it Master's of Science in Data Science now - most of the class designations still carry the old predictive analytics nomenclature in the class codes. I have 3 courses left until I finish so I will be done later this year. It's about half and half these days which classes utilize python and those that require R. The data engineering specialization is almost entirely done in Python. I was managing an engineering team but took a job at a start up that required I manage data engineers and data scientists. I needed to know more about what they do and communicate those results to leadership. I haven't taken a class with Bhatti yet but he is definitely still around. I won't ever really be a data scientist but so far I feel like I am learning enough to have intelligent conversations and ask the right questions of my employees and the models they are building. 

    I love your first two points. That is the one great thing the program reinforces. Data Science should follow a scientific process and there is no perfect model that solves everything you are attempting to do. 

    The start up I work for has decades of cleaned and high quality real estate data. It was all originally built for the founder so he could find off market deals. He realized he could slap a better UI on it, throw some marketing tools in and he has a platform for other real estate investors. Now he wants to build all sorts of analytics capabilities on it to make it more sophisticated and help buyers find deals faster and ultimately make more money. We just started exploring what we can do so who knows where it will end up. 

    • Like 1
  7. 1 minute ago, Tex-19 said:

    Isn't it possible that they're promising multi-year NIL deals that will be signed as soon as the recruit is enrolled? And maybe the $30mm is an endowment amount like Clark Field.

    Idk but it just seems like they've gotta have something planned to keep these kids around. The boosters with that much money can't be quite that dumb. 

    Don’t underestimate aggy stupidity, my dude. Regardless of how rich the individual is.

    • Haha 1
×
×
  • Create New...