Ultimate Chess Puzzle Book rating tests

John Emms published a book titled The Ultimate Chess Puzzle Book.  This question is for anyone who's used the book or has otherwise taken a rating-estimate test from a book.

I'm doing it's tests.  I'm used to most chess rating estimate tests (tactical or otherwise) over rating me, and this book is the same so far (completed 6 out of 15 tests)  Of course all these tests always have the disclaimer at the bottom, take with a grain of salt, but nearly every test I've taken has figured my rating 200-300 points higher than it actually is. 

If I could match an expert or master tactically, I'm sure I would have noticed over the board, but this is not the case, they definitely see more tactics and faster than I do.

They all indicate say my rating is between 200, and 300.

Actual game is more about strategy - puzzles deal only with tactics.

Yes, it's a tactical score -- but in a tactical situation I see less than a master.  Not only less possibilities but I don't see as deep and I can't calculate it as fast... I'm simply outclassed tactically.  Yet the book suggests if I can solve a certain number of problems I have a tactical ability similar to a master which OTB is certainly not the case.

I'm thinking maybe
1) GMs underestimate non-GMs in some respects (masters see more than they realize) and/or
2) He didn't test it on enough people and so the scores are simply wrong for everyone in any case.

I get 1700-1800, which I think is (perhaps incidentally) very accurate, though tactics are not the strong part of my game, so perhaps if you put in some positional stuff you would see the inflation .

Thanks for the feedback Elubas -- I guess it's fair to say tactics are my strong point then (??) -- I guess it's good I'm trying to work on strategy and endgames, I've certainly noticed an improvement in my play since trying to focus on them.

Oh wait, sorry, the first test with 1700 was a while ago, while the second one which I said was 1800, it turns out I started it like a month ago but didn't finish yet, so I still didn't attempt the level 5 puzzles and might still have another level 4 or 2, so I could easily end up scoring 2000+. I tend to stop halfway on these tests and forget to come back to them.

It's hard for any book to accurately measure strength based on how they solve puzzles, simply because there are many more factors than finding the right move after a lot of thought in a puzzle position. I mean if you can solve a 10 move combination, you must be quite good, but it's hard to measure just how good so it's easy to be off by as much as a few hundred points either way.

Still if you're scoring master level (2200+) that's still very good, just not quite  real master strength most likely.

I usually get about 2100. Sometimes 2200. I don't pay much attention to it though.

You're right, there's a lot more to chess, even tactical shots, than finding the first correct move.  I was just curious what other people thought about these tests.

I also don't give it too much credit, to guess my rating I pay more attention to how I do at my local club.  I need to start going to tournaments again... that way I wouldn't have to guess so much