LoL Prediction Script: How Skillshots Hit or Miss

Quick verdict: A LoL prediction script doesn't read minds, it reads movement. It samples enemy waypoints, factors in your projectile speed and cast time, then aims where the target will actually be. Tuning the prediction mode and lead values is what separates a 70% hit rate from a 40% one.

Every skillshot in League of Legends is a math problem. Your target is moving at a known speed along a path the game reports as waypoints. Your projectile has a known speed and cast time. The gap between where you click and where the target actually stands when the projectile arrives is the entire game. A lol prediction script closes that gap by calculating the intercept point in real time, and understanding how it does that is how you stop blaming the script for missed charms.

This isn't a black box. Once you know what the prediction engine is actually measuring, you can tune it per champion and per matchup instead of leaving everything on defaults and hoping. That's the difference between an Emerald Ahri and a Diamond one.

Waypoints: The Movement Data Every Prediction Script Reads

League's engine doesn't report a champion's exact position every frame. It reports waypoints, the click destinations a player has queued up. When you right-click across the lane, the server knows you're heading to that point, and it updates your real position along that line. A prediction script samples those waypoints, extrapolates the path, and aims at where the target will be when your skillshot lands.

Here's where it gets interesting. Good players break their waypoints constantly. They stutter-step, they issue a new click every 200-300 ms, they juke toward the brush and back. A prediction engine that only reads the current waypoint will aim at the brush and miss when the target cuts back. That's why Cado's prediction has multiple modes: waypoint, position, and a hybrid that blends both.

In our testing on the current patch, waypoint mode is the best default for linear skillshots like Morgana Q, Ahri E, and Blitzcrank Q. Position mode is better against targets who spam-click in tight circles, which is common in higher elo. If you watch a Master+ ADC orbwalk, they're reissuing movement commands almost every auto attack. Waypoint prediction on a 1.0 lead against that will over-lead every time.

Why Lead Multiplier Matters More Than Mode

The lead multiplier scales how far ahead of the target the script aims. At 1.0, it assumes the target keeps moving in a straight line at full speed until your projectile arrives. That's correct for a target running away in a straight line. It's wrong for a target who's about to turn, and it's very wrong for a target who's already reacting to your cast animation.

Most players set this too high, especially on slow projectiles. Anivia Q travels slowly, so a 1.0 lead on a moving target often overshoots because the target has time to change direction mid-flight. Drop to 0.85-0.9 on anything slower than about 1400 units per second. For fast skillshots like Ezreal Q or Xerath E, 1.0 is fine.

Projectile Speed and Cast Delay: The Two Numbers That Decide Everything

Every skillshot has a missile speed and a cast time. Morgana Q travels at 1200 units per second with a 0.25-second cast time. That means from the moment you press Q, roughly 0.25 seconds pass before the projectile even exists, then it travels. A target moving at 350 units per second covers about 87 units during the cast alone. The prediction script has to account for both.

Champions with long cast times are the hardest to predict with. Xerath Q has a charge time you control, and his R has a fixed delay before each meteor lands. A good Xerath waypoint prediction setup accounts for that delay per shot, not as a single global value. If you're charging Q for 1.5 seconds, the target has moved 500+ units by the time you release. The script needs to know the charge state, not just the cast.

This is why some shots still miss even with prediction on. If your champion module doesn't expose the charge time to the prediction engine, the engine assumes a full-speed cast and leads short. Check your champion module settings before you blame the script. Most modules have a cast delay field you can tune.

Hitbox Radius: Why Thin Skillshots Feel Unfair

League hitboxes are wider than they look. A Nidalee spear has a hitbox radius of about 35 units around the missile center, which means a target whose model edge clips the spear still gets hit. The prediction engine uses that radius to decide whether a shot is worth firing. If the calculated intercept point is within radius plus a small buffer, it fires. If not, it holds.

That buffer matters. Set it too tight and the script holds shots that would have hit. Set it too loose and you're firing at angles that clip minions or terrain. In our testing, a 15-25 unit buffer is the sweet spot for most linear skillshots. For AoE circles like Orianna Q or Syndra Q, the radius is large enough that buffer barely matters, which is why those champions feel so consistent with a lol prediction script.

Terrain and Minion Collision

Prediction engines don't just track enemy champions. They check the path for minions, terrain, and spell shields. A Blitzcrank Q that would hit a target standing behind a caster minion gets flagged as blocked. Good prediction settings let you toggle minion collision per spell. Turn it on for hooks and binds, turn it off for spells that pass through minions like Ezreal Q.

This is also where evade and dodge logic intersects with prediction. The same waypoint data that helps you aim also helps you dodge. If you're running both, make sure your evade radius and your prediction buffer don't fight each other. We've seen setups where evade pulls the champion out of range mid-cast and cancels the skillshot entirely.

Why Some Shots Still Miss: The Honest Answer

Prediction isn't perfect, and pretending otherwise is how you get frustrated. Here's what actually causes misses with prediction on.

1. The target changed direction after the cast started. No prediction can read a future click. A target who flashes mid-flight or uses a dash will beat any engine.
2. Your cast delay was mis-tuned. If the module doesn't know your Q charges for 1.2 seconds, it leads short by 400 units.
3. The humanizer added a miss. Cado's humanizer can intentionally miss 5-12% of skillshots to look human. If you're in a scrim or a 1v1, drop it to 0.
4. Minion collision blocked the path. A well-placed caster minion eats more skillshots than any dodge.
5. The target has bonus movement speed from a Ghost, a Nami E, or a Chemtech soul. Prediction reads current speed, but speed buffs applied mid-flight change the intercept point.
6. You're using position mode against a stutter-stepper. Switch to waypoint or hybrid.

That last one is the most common. In Diamond+ games, players issue movement commands constantly. A position-based prediction engine sees a target standing still for a frame and fires at that spot. A waypoint engine sees the queued destination and leads correctly. If your hit rate feels inconsistent, check your mode first.

Tuning Prediction Per Champion: Three Real Examples

Generic settings get you to plat. Per-champion tuning gets you to Diamond. Here's how I'd set up three common skillshot champions.

Ahri: Charm and Q Return

Ahri's E has a 0.25-second cast time and a slow projectile. Set lead to 0.9 and enable waypoint prediction. Her Q return is harder because the return path depends on her position when the orb returns, not when it's cast. Most modules handle this correctly, but check that the return prediction is enabled separately. Against a Yasuo who's dashing through minions, no prediction will reliably land charm. That's a matchup reality, not a script flaw.

Ezreal: Q and R Snipes

Ezreal Q is fast enough that 1.0 lead works fine. His R is where prediction shines. The ult has a 1-second cast time and travels across the map. A good prediction engine will lead the target by 3-4 seconds of movement, which is where most players miss manually. Set the R prediction to waypoint mode and increase the lead to 1.05 for long-range ults. Watch for targets who are recalling or channeling, those are free.

Morgana: Q Through Minions

Morgana Q is the classic prediction test. Minion collision matters here. Enable minion blocking and set lead to 0.9. If you're playing her support, the enemy ADC will be stutter-stepping to farm. Waypoint mode reads their last-clicked position, which is usually the minion they're walking toward. That's your intercept point. In our testing, a well-tuned Morgana Q hits about 65-70% of the time against Diamond ADCs. That's plenty.

If you want a broader list of champions that scale well with prediction, our best champions for fast rank up guide covers the top picks by role.

Humanizer and Prediction: Finding the Balance

A perfect prediction script hits every skillshot. That's also the fastest way to get reported. Cado's humanizer adds a configurable miss chance and a reaction delay to your casts. For ranked play, 5-12% miss chance is the range most players settle on. For a 1v1 or a scrim, drop it to 0. The tradeoff is obvious: lower miss chance means more hits, but also more suspicious VODs.

I'd start at 8% and adjust based on your elo. Low-elo lobbies won't notice a 5% miss rate. In Master+, a 5% miss rate on a champion like Xerath is still suspicious because players there don't miss that often. Match your humanizer to your rank, not to your ego.

If you're setting up a fresh install, a well-maintained League of Legends script like Cado ships with sensible prediction defaults, but you still need to tune per champion. The defaults are a starting point, not a finished setup.

Getting Started: A Practical Prediction Setup

Here's the setup I'd use for a mid lane mage on a fresh install.

• Prediction mode: Waypoint for linear skillshots, Hybrid for AoE
• Lead multiplier: 1.0 for fast projectiles, 0.85-0.9 for slow ones
• Cast delay compensation: On, tuned per champion module
• Minion collision: On for hooks and binds, off for pass-through spells
• Hitbox buffer: 15-25 units
• Humanizer miss chance: 8% in ranked, 0% in scrims
• Reaction delay: 120-180 ms, matching your normal reaction time

Load into a practice tool game, turn prediction on, and fire 20 skillshots at a moving target dummy. Count the hits. If you're under 70%, adjust your lead multiplier in 0.05 increments until it stabilizes. Then take it into a normal game and see how it feels against real players. The practice tool doesn't stutter-step, so your real hit rate will be lower. That's normal.

Prediction is the single biggest quality-of-life feature in a LoL script. It's also the one that rewards tuning the most. Spend 15 minutes per champion in the practice tool and you'll feel the difference in your first ranked game. If you want to see how the rest of the script fits together, the full LoL script guide covers orbwalker, evade, and combo settings in the same level of detail.

FAQ

Why does my prediction script still miss skillshots?

The most common causes are a mis-tuned cast delay, the wrong prediction mode (position instead of waypoint), or a humanizer miss chance that's set too high. Targets who dash or flash mid-flight will also beat any prediction engine, that's a game mechanic, not a script issue.

What lead multiplier should I use for slow skillshots?

Drop to 0.85-0.9 for anything slower than about 1400 units per second. Slow projectiles like Anivia Q or Morgana Q give the target time to change direction, so a full 1.0 lead often overshoots.

Does a lol prediction script work on all champions?

It works on any champion with a skillshot, but the quality depends on the champion module. Champions with charge times or multi-stage casts (Xerath Q, Varus Q, Pyke Q) need modules that expose those timings to the prediction engine. Most well-maintained scripts include them.

Can prediction be detected or reported?

Prediction itself is just math on visible movement data. What gets players reported is impossible accuracy. Keep the humanizer miss chance at 5-12% in ranked and you'll look like a good player, not a script.

How do I test my prediction settings?

Practice tool with a moving target dummy. Fire 20 skillshots, count hits, adjust lead in 0.05 increments. Real players stutter-step, so expect your in-game hit rate to be 10-15% lower than practice tool.