Why Facebook adverts are not working

Some months back I reviewed FaceBook’s sponsored listings  and found the Click through rate to be very poor. Since then I have tried with a much larger brand and had an even more disastrous CTR. Today I think I found out at least one reason why.

I was going through Facebook and was clicking around. Just as I clicked, I noticed an advert saying “let’s play chess…” but by then I had clicked somewhere else. The advert that replaced the chess one was totally different. I WANTED to get to the chess advert. I started searching… but no joy. Someone had an advert targeted at me and it passed me by before I could click. It might be that I see it again, but there is a problem with the algorithm that dispalys the adverts. FaceBook should try having the advert display for about 5 pages before rotating, to ensure it gets into the “head” of the Facebook user if they really want people to start clicking.

Until they do that, or find other solutions, we are finding unacceptably low click through Rates. Mind you – since I am paying on a CPC I don’t care… but Facebook needs to fix that to generate income.

E3 is a must in the Nimzo-Indian

I am learning very late in life that you need to learn the theory, not just think for yourself. Case in point on the ICC yesterday when I played a set of moves I always thought was obvious as white: 1.d4  e6 2: c4  Nf6 3:Nc3 Bb4 4:Bd2 d5 (Diagram below).  

nimzo-indian-move-4

Well – Dasher’s crafty had already been suggesting e3 instead of Bd2 on the previous move, but with a score of 0.01 at the moment it looks OK. That score, though, was assuming that I did do e3 this time round. Ignoring e3 now was an immediate disaster. I chose 5: cXd5 and guess what crafty thought of that?

nimzo-indian-move-5

Well – after 58.1 Million nodes tested, it gave a bit of a no-no. But if you are like me and a very average human chess player, the situation is much worse. Crafty was still saying -0.5 until it got thought that far down. Either way, not doing e3 was the start of a slippery slope. I lost.

Moral for the average player (like me): Develop at the opening instead of sacking things half-cock.

How to Analyse Chess Games

I’m really no expert at Chess, but I think I have figured out how to go about analyzing my online chess games after the event. Reading games in newspapers is one thing, but it hurts my brain to be honest. But by using “Dasher” to play my games on the ICC it is really quite easy to see after the game where I went wrong. Dasher is one of the down-loadable clients that is used to play chess against other players there. It comes with “Crafty” built in – so you can play chess with a machine when offline. You can’t use the computer to play other people (at least I HOPE you can’t, that would be cheating!) but the system saves all your games for you. You can then select a game after the fact and bring up a few screens that will help you see very quickly where you went right (or where your opponent went wrong). I also have Fritz on a computer somewhere, but that doesn’t talk so easily with the ICC and that would mean I would probably have to import any game for analysis. Fritz does, however,  come with a graph of your score during a game, which makes it blindingly obvious when you cocked things up if you are a poor standard like me. Dasher, though, is just as easy to interpret once you know what it all means.

Here is a screenshot of a recent opening.

How to analyse Chess Games

It is me (white) to move and I am on move 4. The chess engine, “Crafty” shows the moves that I could realistically have made at this point. Nxd4 is shown as 66% (the red oval) with Bc4 only 26% or c3 at 8%. Now – I played Bc4 (Green oval) instead of Nxd4. I seem to have gone wrong on move 4. To  fair, that does not suggest I have gone wrong, but does show what the book moves generally are and unless you are WAY better than me, the book moves are all you need to learn for now.

So… 30+ years of playing chess an I can see that on a major line (Scotch gambit: Vitzhum attack according to Crafty), I am going off piste on my fourth move. Wow I have so much to learn!

But it gets worse just one move later. Until that point the “score” was marked as zero. that means that I had just done unusual moves, not bad ones. Look what it says after my next move:

How I can see I made a bad move

Now it is actually analysing my opponent’s next move, but in the process it has reduced white’s score from 0.00 (which means even or “in book” to -0.22. Now, anything more than -1 means I am about to go a pawn down, so I should go back and see why moving that knight to g5 was such a pants move. Looking at it now it’s obvious. You don’t start moving pieces a second time unless you’ve got all your pieces developed and your castle’s centralised or if you have a real good reason.

So that’s how I’m going to start to analyse my games. For the record, I lost the game a few moves later, even though I hung on until move 26, when my opponent forked my king and rook with a knight.