Almost everyone who trains memory hits a memory training plateau. The first few weeks are exciting: scores climb from session to session as the techniques click. Then, one day, they stop. You run the same drill for weeks and land on roughly the same number. This is normal, and it does not mean you have reached your limit. It usually means one part of your process has become the bottleneck, and doing more of the same drill will not fix it. This guide shows you how to find that bottleneck and what to do about it.
First, check that it is really a plateau
Scores in memory sport are noisy. A good sheet, a bad night's sleep or a lucky deck can move a result a long way. Before deciding you are stuck, look at the trend rather than your best score.
- Use an average. Take the average of your last five attempts in the same format and compare it with the average of the five before. A rising average is progress, even if your personal best has not moved.
- Compare like with like. Only compare runs with the same number of items and the same time limits.
- Look at accuracy separately. A flat score with rising accuracy is progress too, because accuracy is usually what unlocks the next jump in speed.
If the average has genuinely been flat for three or four weeks, read on.
Why plateaus happen
Every attempt is made of several steps, and your time or score is limited by the slowest one. Early on, everything improves together. Later, the easy gains are gone and one step is holding the others back. Repeating full drills works that step only a little, and all the others a lot, so progress stalls.
For most events the steps are the same:
- Recognition: seeing the digits, card or word and knowing what it is.
- Conversion: turning it into your image.
- Placement: putting the image in the right location and moving on.
- Retention: keeping it there until recall.
- Recall: walking the route and turning images back into answers.
The fix for a plateau is to find which step is slowest, and train it on its own.
How to find your bottleneck
Test conversion alone
Take a set of flashcards or a random list of items and time yourself converting each one to its image, with no palace and no recall. If you hesitate on some images, or the whole list is slow, conversion is a likely bottleneck. This is especially common after learning a new system such as the Major System or PAO, where a handful of weak images slows everything down.
Test placement alone
Walk your route mentally as fast as you can while still seeing each location clearly. Then walk it while placing a simple, familiar image at each point. If the walk is slow or you lose your place, the route is the bottleneck.
Look at your mistakes
The type of error tells you a lot. Wrong images point to conversion. Right images in the wrong order point to placement. Blanks point to retention. A cluster of errors at the same point in several attempts points to a weak part of the route. Our guide to reading your memory drill results sorts errors into types in detail.
Compare memorization and recall
If you often know that an item was memorized but cannot bring it back, recall is the issue, not memorization. That calls for stronger images and a calmer recall routine, not more speed.
Fixes for each bottleneck
Slow conversion
- Find your slow images. Time yourself on each image and list the ones that take noticeably longer. They are usually the same few every time.
- Replace, do not just repeat. An image that is always slow is often a poor image. Swap it for something more vivid or more personal.
- Drill the slow ones daily. A few minutes a day on the weakest images is more useful than going through the full list.
- Use spaced review. Let the images you know drift further apart and bring the hard ones back often. This is how the MemoChamp Review page schedules your own images, and the idea is explained in spaced repetition.
Slow or unreliable placement
- Walk your routes more often. A smooth route is walked without thinking.
- Replace weak locations. Two similar locations next to each other are a common hidden cause of errors.
- Try a different number of images per location. Moving from one to two images per locus shortens the route, but only helps if you can keep them in order. See how many memory palaces you need.
- Rotate your routes. Reusing a route too soon leaves old images behind and causes placement errors.
Poor retention
- Make images interact with the location. An image floating near a location is easy to lose. One that smashes, soaks or melts part of the location stays.
- Exaggerate. Size, movement and absurdity all help. The visualization guide covers what makes an image last.
- Slow down slightly. A fraction of a second more per image can remove a whole category of blanks.
Recall problems
- Recall in a fixed order. Walk the route from the start, every time.
- Do not stop at a blank. Move on, finish the route, then come back. The context of neighbouring images often brings the missing one back.
- Guess where there is no penalty. In events without negative marking, a blank is never better than a guess.
Change how you practise, not just how much
Stretch runs and consolidation runs
Alternate two kinds of drill. A stretch run is slightly faster or longer than you can comfortably manage, which shows you where things break. A consolidation run is at a pace you can do cleanly, which builds reliability. Doing only consolidation runs keeps you where you are. Doing only stretch runs builds error habits.
Accept a temporary dip
When you push speed, accuracy usually falls for a while before it recovers at the new pace. That dip is part of the process. Judge a change after two or three weeks, not after two or three runs.
Train parts, not just wholes
A full drill tests every step at once. Spend some sessions on the weak step alone: image flashcards for conversion, route walks for placement, or small sets with long delays for retention.
Change one variable at a time
If you change your images, your route and your pace in the same week, you cannot tell which change helped. Keep a short note with each session saying what you changed.
Take a lighter week
Tiredness accumulates. If your scores have drifted down rather than stayed flat, a lighter week with more review and fewer drills can help as much as anything else.
Plateaus by event
- Speed Cards: often conversion. Card images that are nearly automatic are not automatic enough at speed. Accuracy matters most, because an incomplete deck scores next to nothing. See why a perfect Speed Cards deck beats a fast one.
- Speed Numbers: often placement or the first rows. One slip halves a row, so look for rows lost to one or two mistakes.
- Random Words: often retention of abstract words. Build a stock of go-to images for words that are hard to picture.
- Names and Faces: often the link between the name image and the face. Pick a clear feature before thinking about the name.
- Hour events: often stamina and route length rather than technique. See training for the hour events.
A four-week plan for getting unstuck
- Week one: measure. Run your usual drills, record your average and classify every error.
- Week two: fix the biggest step. Spend half of each session training the bottleneck on its own, and keep a couple of full drills.
- Week three: stretch. Add stretch runs slightly beyond your comfortable pace, alternating with clean consolidation runs.
- Week four: re-measure. Go back to your usual drills and compare the five-run average with week one.
If the average has moved, keep going with the same approach. If it has not, the bottleneck was somewhere else: go back to week one and look again.
Keep perspective
A plateau is a sign that you have used up the easy gains and need to train more precisely. Every experienced memory athlete has been through many of them. Measure honestly, find the slowest step, work on it alone, and your scores will start to move. When you are ready, pick a drill in training and start with a measurement week.
