So I make it no secret that I use AI pretty consistently in my day-to-day life and it has transformed my efficiency in a lot of things (for example, it helped me re-furnish our entire 100-year old piano, as well as teach me how to replace screens on windows). I also use it for work, allowing me to work a second full-time job in about 5-7 hours a week (marketing FTW).
I am a firm believer that it is a
tool that needs to be used correctly. Heck, even my doctors at work are using AI in their patient rooms to pull together their conversation into doctor's notes. It's the future, but only if used appropriately.
I decided at the beginning of the offseason that I wanted to come into next season with a very specific "strategy" focusing on building a roster around players with a specific offensive and defensive playbook. For my team, I chose I wanted to run the Bills 08 Defensive Playbook next season, for better or worse.
This is an example of a positional analysis I ran this morning for outside linebackers (OLBs) and what the AI gave me. I share this one as I have zero intent to draft OLBs in the upcoming draft.
Mind you, I've worked a lot of time to get my ChatGPT pro account to deliver information to me in a way that I like and I've trained it on a lot of things.
OUTSIDE LINEBACKER (OLB) – FULL CPU-vs-CPU ANALYSIS (Madden NFL 0
Environment Assumptions
• CPU vs CPU only
• Madden NFL 08
• Bills Defensive Playbook
• INT Slider = 5
• Injury Slider = 102
• Emphasis on discipline, containment, and tackling (not splash plays)
No OVR. No hidden traits. No modern attributes.
1. What OLB Actually Does in Madden 08 CPU Sims
In a Bills defense, OLBs are
not edge rushers and
not playmakers.
They are:
• Edge setters
• Flat / curl zone anchors
• Run-support stabilizers
• First line of failure prevention on outside plays
When OLBs fail, you don’t lose sacks — you lose:
• 6–9 yard runs to the edge
• TE dump-offs converting 3rd-and-medium
• RBs consistently falling forward
That’s why “boring” OLBs win CPU games.
2. Common OLB Failure Modes (Why the Model Looks Conservative)
What kills Bills defenses:
• Over-pursuit
• Missed tackles in space
• Late reactions to outside runs
• Poor underneath zone spacing
What does NOT matter much:
• Sack totals
• Highlight hits
• Straight-line speed without awareness
This model minimizes defensive leakage, not stat chasing.
3. OLB Scoring Formula
OLB Score =
(AWR × 0.2
+ (TAK × 0.22)
+ (STR × 0.12)
+ (STA × 0.10)
+ (INJ × 0.10)
+ (SPD × 0.10)
+ (ACC × 0.0
4. Attribute-by-Attribute Breakdown
AWR (28%) – Assignment Integrity
• Zone depth discipline
• Reaction to play-action
• Containment responsibility
Low AWR causes silent defensive failure:
• TE leaks
• RBs slipping into flats
• Chain-moving completions
TAK (22%) – Drive Control
• INTs are suppressed
• Drives end via tackles
• OLBs are often first to contact
Low TAK turns 3-yard gains into 7-yard gains.
STR (12%) – Edge Authority
• Taking on TEs
• Preventing seal blocks
• Eliminating cutback lanes
Weak OLBs get washed out.
STA (10%) – Late-Drive Stability
• OLBs rarely sub out
• Fatigue degrades AWR and TAK
• Accelerated clock amplifies this
INJ (10%) – Structural Continuity
• Injury slider at 102
• Losing an OLB breaks run fits across the defense
Availability is a competitive trait.
SPD (10%) – Containment Range
• Outside runs
• Stretch plays
• Chasing plays back inside
Speed never overrides awareness.
ACC (8%) – First-Step Reaction
• Closing on flats
• Preventing easy dump-offs
• Separator, not a driver
5. Explicitly De-Emphasized Attributes
These are intentionally excluded:
• AGI
• BTK
• JMP
• CAR
• TGH
They add noise, not value, for CPU OLB play.
6. Sanity Filters (Strongly Recommended)
Before scoring:
• AWR < 65 → Bills-incompatible
• TAK < 70 → Drive-extension liability
• INJ < 75 → High-risk depth only
Fail two of these and the player is not draftable in this system.
7. How to Interpret OLB Scores
• High score + low speed → Excellent Bills OLB
• High speed + low AWR → Trap player
• Balanced profile → Ideal OLB2 / OLB3
• Low STA → Rotation only
8. One-Sentence Truth
Bills OLBs don’t make plays — they prevent bad ones.
That is why this model looks conservative — and why it wins in CPU simulations.
