ARRL 10m Contest 1998

Normalised Scoring: Introduction

Contesters are often urged to set their own standards, to compete against their own benchmarks, and not to see top-ten listings as the only measure of contesting success. I agree. But it is difficult to set those personal objectives in the ARRL 10m contest. More than any other HF contest, absolute performance here depends very heavily on the solar cycle. Not only does total score vary enormously, but so do the differentials in scoring expected through geographical advantages (or disadvantages). And worse still, most people are fair-weather contesters when it comes to ten: few will compete in years when QSO totals reach only a few hundreds, so volumes of competitors vary a lot too. Knowing how well a score ranks relative to any fixed standard is therefore virtually impossible. For someone like myself who ...

• enters every year
• resides at a moderately high latitude in the northern hemisphere (you have to be in the southern to do well in all but the highest spot years)
• has a long series of country 1st place certificates won mostly through lack of any serious local competition
• is competitive and wants to measure up against his peers

... there is now a solution. The 1998 contest got record numbers of entrants (well the sun is a lot spottier now so more of us creep out of the LF woodwork) and this allows a chance of trying out realistic score normalising factors.

The idea of normalising is to remove differentials due to different point scoring regimes, transmitter power level, and the availability of stations and multipliers due to geography and propagation. No normalising scheme is perfect, and I admit that the method used here contains flaws. However, I tried several variants before I was happy enough to put this down in bytes. Its the best I could come up with in the time available. A discussion of the methodology will make comforting reading for those whose scores didn't make top places in the normalised listings!

The points you must know before peeking at the scores are that high power stations, and multi-operator stations especially, are a little under represented in the normalised top score lists. Also certain geographical areas (e.g. W2) have slightly over-enhanced normalised scores. The distribution of mixed-mode v phone v CW is about right. There are two sets of scores.

• Normalised Scores Listed by Class and Power. Scores are listed by entry class (e.g "mixed mode, low power") and compare on "equal" footing all entrants in that class regardless of geography.

• Normalised Scores Listed by Area. Scores are listed by area (continent, US call area) and compare on "equal" footing everyone in an area regardless of power or mode(s) used.

The normalised listings are restricted to the top 5% in each category. Anyone who isn't listed here and is curious enough to see his or her relative placing, a mail message to g0aev@explore.force9.co.uk will get you the info.