To the memory of the memory of Bahrain

I had a dream last night that I was competing in a muffin-making competition. Not baking the muffin itself, you understand, taking it out of the packet, toasting it in the grill and then spreading butter and jam on it. I was quite confident of success, but I woke up before it finished, so we'll never know.

Perhaps slightly less exciting, but still worth talking about anyway, let's see what I can remember at this late stage of the rest of the World Memory Championship. The start of day two was delayed somewhat, waiting for Tony Buzan to arrive and announce the scores so far. I don't know why it's essential for Tony to perform this task - he does have by far the best speaking voice, but it's just a case of reading from a piece of paper, and I'm sure there are other people among the WMSC who are capable of doing that. Phil did his best to hide his irritation with the Chinese translator, who kept pointing out that the schedule said to start at eight, it was now after eight, and we should have started by now.

When Tony did turn up (perhaps distracted by his sleeping arrangements - I had by now been moved from my suite into a normal room, but apparently Tony, who was staying in the posher and more expensive Gulf Hotel, had had to put up with a double room instead of the suite he'd booked for the first night), I was pleased to see that he'd corrected his description of Andi - he'd referred to him all through the previous day as 'two times world champion Andi Bell' instead of three times champion, but it seems that was a genuine mistake rather than any kind of political title-stripping, and he described him correctly from this point onwards. I always had my name read out as 'reigning world number one Ben Pridmore', and Gunther as 'reigning world champion Dr Gunther Karsten', by the way.

Anyway, having confirmed that, as I said at the end of my last-but-one blog entry, I was 200 points up on Gunther, 500 on Hannes and 550 on Chuanwei, I was feeling confident. Hannes sounded pessimistic when he told me he couldn't see any way he could make up that difference, but I reassured him that there's always the possibility of me messing things up. Secretly, I didn't believe he could make up that gap either, but my opinion was soon to change.

The first event of the day was Speed Numbers. Which, for the sake of consistency with other disciplines, we really ought to rename Five Minute Numbers, but that's just me being pedantic. Anyway, it's exactly like Hour Numbers except (as you might have guessed by now), there's just five minutes of memorisation time. Since, once again, the numbers are in rows of 40 and you drop 20 points for a single digit wrong in a row, and score nothing for a row with two or more errors, this means that an error strips you of a bigger proportion of your points than in hour numbers, and there's a bigger chance of a disaster. Still, with a smaller quantity of digits to memorise, there's also a big chance of getting much more than a twelfth of the score you can achieve in the 60-minute discipline.

This was another discipline where I thought I had a chance of beating the world record. Andi's 396, set in Germany last year, was huge, but with him not on top form and me knowing I could consistently get 360 (better than anyone else had managed) and 400 at a stretch, I was quite hopeful. You get two trials at Speed Numbers, with the best score of the two counting, so normal procedure is to do a 'safe' attempt first time, then if that goes well, try for something huge and run the risk of messing it up. My 'safe' score is 360, and I attempted that, but, annoyingly, I blanked out completely on one of the images. Luckily, it was in the last row, and you're allowed to partially complete the last row you attempt, so I just scribbled out everything after that point, and ended up with a score of 336. Pretty decent, all in all.

We get the results of the first trial before we start the second (lots of frantic marking by those wonderful arbiters!), and that confirmed that my 336 was the best score. But only by three digits - Hannes had 333. Andi had got 278, followed by ever-consistent Guo Chuanwei with 240 and Liu Ping with 224. Gunther had only got 97. Still, all that counts for nothing if anyone gets a better score in the second trial, so I went for something big. Something big for me is 468 digits (I put 234 on each journey - a strange number, I admit, but that's just how it works out for me; 26 locations on each journey, three images in each location, three digits to an image), but it went wrong in a new and unexpected way.

As I said above, I memorise the digits in groups of nine, and when I got to the end of the ninth row, I somehow moved back to the start of row nine, instead of row ten, and memorised all that row of digits again (different images, naturally, since I do them in threes, which means they overlap the ends of the row, so while the first time it was split --1/234/567/890, second time it was 123/456/789 etc.) I've never done that before, and I didn't notice it until I got to the end of what should have been row 11, and noticed it was labelled 'row 10'. The resulting mental confusion put paid to any chance I had of beating 336.

I don't actually remember whether we got the results of trial two before we moved on to the next discipline. I'm going to assume we didn't, because we don't usually, but I can't be certain. It wouldn't have made much difference to my state of mind, because next up was Historic Dates, and I knew in advance what was going to happen there. In this discipline, you get a list of years (between 1000 and 2099 - the full title is Historic And Future Dates, but not many people use that) and a description of something fictional that happened in that year. You get five minutes to memorise them, then you get the list of happenings back in a different order and have to fill in the years beside them. The number of dates provided has just been increased this year from 110 to 120, so there was another good chance of a world record here.

I always used to be the best in the world at dates. And the best by miles and miles, so much so that it was embarrassing. However, over the last couple of years, Hannes has become just a little bit better at it, and always narrowly beats me. It's really annoying - now there are two of us who are light-years ahead of the rest, but I'm always in second place. Nonetheless, I gave it my best shot. I used to always just about get to the end of the 110 with ten seconds to spare in five minutes, so I was a little surprised to find that I got all through the 120 this time, and still had about ten seconds to spare. I remembered most of them, too, and estimated that I'd got a score of about 100 (one point for a correct date, minus half a point for an incorrect one, no score for a year left blank). The world record was 99 (set by Hannes last year - my best was 96).

It would have been around this point, probably, that we got the results of the Speed Numbers. Much to my surprise, Hannes had set a new world record of 405! I had no idea he was capable of that, and it caused me to reconsider my chances of winning the championship. My 336 was still the second-best score, but Ping had moved into third place with a 300, and Gunther had done 280. This made the top ten at the half-way point in the championship look like this:

1 Ben Pridmore 3749
2 Johannes Mallow 3425
3 Dr Gunther Karsten 3390
4 Chuanwei Guo 2942
5 Andi Bell 2783
6 Boris Konrad 2689
7 Su Ruiqiao 2678
8 Yip Swe Chooi 2476
9 Liu Ping 2408
10 Yuan Wenkui 2387

A healthy lead, it might look, but I was getting worried. Johannes, as I mentioned, always makes up a bit of ground on me at historic dates, and he's also always better than me at the next discipline, Random Words, even when I'm at my best. And words is a discipline where there's always a good chance of me making a mess of things and losing a LOT of championship points. It was going to be closer than I was comfortable with.

Still, I was in reflective mood - stepping out of the building for a blast of the oven-like atmosphere of Manama (it was a little cloudy, but still unbearably hot), my thoughts were mainly running along the lines of "awww, we're six-tenths of the way through the competition, and then it's a whole year till the next. It goes by so fast!"

Still, it was already time to go back inside to the air-conditioned hall for the next discipline. The temperature, incidentally, was as never-quite-right as in every building with air conditioning - the first morning, it was decidedly chilly in there and I was regretting not bringing a jumper. For the rest of the event the temperature was about right, but there's always that problem of how, as soon as you turn the aircon off for a minute, the temperature goes up a little, but it immediately feels stuffy, humid and uncomfortable as the outside air gets in. All in all, though, there weren't any serious temperature problems - unless you're Ytep Ho Yin, who spent the whole competition dressed in heavy coat and scarf.

Random Words essentially does what it says on the tin - you get a list of random words and fifteen minutes to memorise them. They're arranged in columns of twenty, with the usual scoring - 20 points for a complete column, 10 points for a column with one mistake, zero for two or more mistakes. I have a habit of having blanks in lots of columns and getting a very mediocre score. I felt a bit under pressure this time, which is usually a good thing, because I always perform best when I absolutely have to. I went for a big 200 words, and felt sort-of more-or-less confident of getting most of them right. My problem is often that I remember a different word that means the same as the one written down - especially if the synonym is one that appears on my list of mental images for cards and numbers. I did feel proud of myself for remembering that one of the words was 'vulture' when my brain was telling me 'eagle', though. 200 would be a great score, maybe better than Hannes and anyone else (with the obvious exception of Boris, who always gets a huge score at words).

Then it was lunchtime, nearly. Sandwiches and things were laid out on the tables, wrapped in cling film, but the catering staff wouldn't let us touch them until the scheduled time of one o'clock, so we had to wait around for fifteen minutes.

After lunch, it was the results of the dates - as per usual, flipping Johannes Mallow had beaten me into second place. My world-record-beating 100.5 looked meagre in comparison to his 110.5; the third-best score was Boris's 69.5. So now we were looking at a championship position like this:

1 Ben Pridmore 4854
2 Johannes Mallow 4641
3 Dr Gunther Karsten 4072
4 Chuanwei Guo 3476
5 Boris Konrad 3454
6 Su Ruiqiao 3398
7 Andi Bell 3217
8 Liu Ping 3029
9 Yip Swe Chooi 2911
10 Yuan Wenkui 2750

Yikes, that lead's coming down. I could really see myself losing at this point - if I'd got 50 fewer words than Hannes (by no means impossible), it was dead level. The final three disciplines are ones in which I was fairly sure I still had the edge over him, but look at what happened last year in hour cards AND speed cards. That had lost me the 2007 championship, so who's to say it couldn't happen again.

Just outside the top ten, incidentally, my compatriots Ed Cooke and James Ponder were fighting a thrilling battle - they'd only been separated by a few points all the way through the championship. Ed at this point was 11th with 2603, with James on 2514. With $1000 on offer for tenth place and nothing for eleventh, it was an important struggle, too.

So we settled down for an afternoon of Hour Cards - sixty minutes to remember as many packs of cards as possible. By this point it was really starting to bother me that I hadn't broken any world records yet. I ALWAYS set at least one new world record at the world championships! I'd got personal bests in three of the first six disciplines, but not in any of the ones where I hold the record. What should I do in the cards? Last year I attempted 36 packs and had a horrible, televised disaster (they'd shown The Mentalists on the big screen during lunch on the first day, just in case anyone didn't know what had happened, so everyone was talking about it). I knew I was on better form this year, but decided to play it comparatively safe and go for 33. My world record was 27, and that gave me a good chance of beating it.

I bring my own cards to competitions these days (they do get shuffled by someone else before I get to memorise them), but those who don't might have been surprised to see that they've got a new supplier this year - some American company, whose cards look decidedly cheaper and flimsier. I don't think the new ones are going to be reusable more than a couple of times. Still, that's just me nitpicking, it didn't affect me at all.

My memorisation went fairly well - I had more gaps and guesses than I was hoping for, but it wasn't a total disaster. I estimated that I'd got about 25 packs, maybe more, maybe less. That should be enough, right? Even so, it was in worried mood that I left the conference centre in the evening. I could just see the headlines - "Pridmore Beaten By Third Successive German In World Championships", "What A Big Fat Loser", and so on. When I'm worried about losing a memory championship, I generally assume the newspapers will focus on me and my weight problem rather than on the winner. I'm self-centred that way.

Tune in later to see whether I won or not! And no peeking at the results!