So yesterday I slid into another Comet in Heyd for a fight and it turned out to be Johnny Nailbomb. So we had a good fight and he ended up winning the fight, barely. You can see the clip for yourself here. After the fight one of my corp mates mentioned that it had been streamed live, so I went over and said hello on the stream.
Fellow Stay Frosty pilot Phoenix slid into the plex after I exploded and took care of his Comet. So everything evens out in the end. Some people in chat said it was unfair to double up on him, but that is not what happened. Our fight was already over. Although to be clear, this was not an agreed upon 1v1. Just a random fight. So even if Phoenix had arrived earlier it would have been fine. Either way, just another day in space. You win some, you lose some.
Sliding into a plex with someone already inside is inherently challenging. The bonus always goes to the defender. In most cases they have the upper hand. And this fight was no exception, he already had his drones out and got the chance to shoot first. I've said it a million times, but in a close fight whoever shoots first has a distinct advantage. That is fair by the way and is a hill I have to climb every time I do this. It is all just part of the challenge.
I was curious if I could take the log data from the fight and feed it into ChatGPT for an analysis. So I copied the data from the fight and asked it to run an analysis of the fight. It is important to keep in mind that it is based entirely on the log, so it makes some interesting assumptions. But here is what I got back.
______________________________
Engagement Window: 21:46:26–21:47:00 (~34 seconds)
Your ship: Federation Navy Comet
Enemy: Johnny Nailbomb (Federation Navy Comet)
✅ 1. Damage You Dealt
Your Railgun Damage (150mm Railgun II)
Number of railgun hits you scored: ~17 hits
Hit quality included: grazes, hits, penetrates, smashes
Total railgun damage YOU dealt:
➡️ ≈ 2,230 damage
Rough breakdown:
- Several strong hits: 170, 196, 192, 174, 150, 165, etc.
- A few glances or grazes: 63, 109, etc.
- You fired consistently with only one “miss completely.”
Your average railgun hit on target:
➡️ ~130–150 damage per shot
(which is very good for a Comet vs. Comet fight)
Your Drone Damage (Hobgoblin II)
You scored: ~20 drone hits
Damage range: 18–60
A couple misses, but overall solid.
Total drone damage YOU dealt:
➡️ ≈ 540–600 damage
šÆ Total Damage YOU Did:
➡️ Approx. 2,750–2,800 total damage dealt
This is enough to kill a Comet, assuming no reps or beefy tank.
❌ 2. Damage You Took
Enemy Railgun Damage
Enemy hit frequency: very high, almost every 1–2 seconds
Enemy railgun hits ranged:
- Low: 96, 101, 122
- High: 204, 193, 192
Total railgun damage taken:
➡️ ≈ 2,900+ damage
Enemy had excellent tracking and hit quality, including:
- Multiple penetrates
- Several smashes
- Very few grazes
- No misses
Enemy Drone Damage
Enemy drones hit A LOT early in the fight.
Damage examples: 40, 55, 57, 50, 63, etc.
Total drone damage taken:
➡️ ≈ 650–700 damage
(You started shooting them later in the fight, reducing their pressure.)
š„ Total Damage YOU Took:
➡️ Approx. 3,600–3,700 total damage taken
This exceeds the typical buffer + rep cycle of a Comet unless you were double-repping or fully charged.
⚡ 3. Capacitor Analysis (VERY Important)
Multiple warnings:
- 21:46:54: Afterburner cannot activate (only 2.1 GJ cap left)
- 21:46:55: Warp disruptor cannot activate
- 21:46:58: Ancillary armor rep cannot activate (only 3.8 GJ cap)
Also:
Enemy NOS is draining you every few seconds, even though it says 0 GJ (because your cap is lower than his).
Conclusion:
➡️ You fought the entire second half of the engagement nearly cap-dry, losing:
- propulsion
- pointing
- your ancillary rep
- possibly gun cycle time
- tracking bonus from afterburner speed
This dramatically reduced your tank and ability to control range.
šÆ 4. Hit Quality Comparison
Your Hit Quality
- Mostly Hits and Penetrates
- Several Smashes
- Only one railgun miss
- Occasional drone misses
➡️ Your accuracy was good, but not perfect.
Enemy Hit Quality
- Large number of Penetrates and Smashes
- Very few grazes
- No misses except drones
➡️ Enemy's accuracy was excellent — likely inside optimal + tracking bonus + maybe web advantage.
š°️ 5. Drone War Analysis
You damaged his drones late but:
- For the first 10–12 seconds, his drones dealt heavy pressure
- Your drones applied consistently but never achieved a wrecking blow
- He did score a "Wrecks" hit with a Hobgoblin (129!)
You eventually ADS your railguns + drones to start removing his drones, but you started a bit late.
š„ 6. Engagement Flow Summary
Early Fight (21:46:26–21:46:33)
- Enemy opens strong
- You miss your first shot
- His drones and guns hit hard
- You get some solid hits back
- Cap starts draining immediately
Mid-Fight (21:46:33–21:46:45)
- You’re landing strong volleys
- Enemy volleys stronger and more accurate
- Your cap is nearly empty
- You start losing rep cycles + speed
Late Fight (21:46:45–21:47:00)
- You begin shooting enemy drones (good timing)
- Your tank collapses due to cap starvation
- Enemy maintains full damage output
- You lose propulsion and tackle
- Recharge too low to activate rep
š§¾ 7. Key Trends Identified
✔️ Your strengths
- Very strong railgun volleys (some >190)
- Good drone application
- Nearly perfect railgun uptime
- You dealt kill-level damage
❌ Your weaknesses
- Capacitor death spiral due to enemy NOS
- Repeated failed module activations = tank collapse
- Enemy had better tracking or range control
- Enemy drones applied heavy damage early
- Enemy guns hit more consistently and harder
- You started killing drones a bit late
š§ 8. Strategic Takeaways
1. You died due to cap starvation, not DPS failure.
Your DPS was 100% enough to kill him.
You lost because you ran out of cap, losing rep, propulsion, and tackle.
2. Enemy appears to have:
- Better range control
- Better tracking
- Possibly:
- Dual webs
- A tighter orbit
- Higher skills (Sharpshooter V, Small Hybrid V, etc.)
- Or he simply had better heat usage
3. You might have been scrammed + single-web while he was dual-web.
That would explain:
- His tracking dominance
- Your inability to get under his guns
- His low number of grazing hits
4. Killing drones earlier may have changed the fight
His drones dealt 650–700 damage — that is a LOT.
Without that, the fight becomes nearly even.
5. Your rep failed due to cap issues, not timing.
Ancillary rep not firing = death sentence.
⭐ Final Verdict
You dealt: ≈ 2,800 damage
You took: ≈ 3,600+ damage
You actually performed VERY well offensively — your output was competitive and at times superior.
But:
⚠️ The enemy NOS + your cap instability swung the fight completely.
⚠️ His tracking + range control gave him more effective DPS.
⚠️ His drones applied better early pressure.
If cap had held for even 10 more seconds, this fight could have gone the other way.
_____________________________
This analysis is fascinating. It is also inherently wrong on several fronts. But I do believe it is interesting and helpful in some ways, as long as you go into this with actual knowledge about how Eve works. To be fair, like I said above, ChatGPT was operating on just the log data and made a bunch of assumptions that turned out to not be true.
I have no idea where it got that I was shooting drones. I never did shoot his drones.
It spent a lot of time worrying about my cap, which was not an issue. I didn't have any cap issues and was outside of his NOS range the entire fight. He also did not have dual webs, his tracking was better because he was using a Magnetic Field Stabilizer in his lows. It's as simple as that. Otherwise our fits were almost identical. The fight came down to two factors, which ChatGPT did manage to nail in the analysis - 1) Drones already out, and 2) the effect of the Mag Stab on hit quality.
Now I typically mitigate those early advantages with some advantages of my own - including superior hull tank and piloting skill. Most of the time that works. But in this instance I was unable to mitigate the early advantage enough. Simple as that.
Anyway, I think using AI to analysis your fights could be advantageous if you already understand enough about Eve mechanics, but I wouldn't recommend it for new or inexperienced players - at least not right now. If you read that the wrong way you might come to the wrong conclusions and just end up confusing yourself more.
So all in all I believe it is an interesting tool that can only improve over time. But right now, it can be confusing and inherently wrong about important elements of the fight.
Cheers.
Comments