2018 female prediction rankings

blob · All-time great

RE: 2018 female prediction rankings

blob · All-time great

Something interesting to note here, is that, because of the end date of the competition, points scored by players after the week of the YE Finals are efffectively locked in - they won't come up to be removed from a players ranking score until after the next years forum competition is complete.

This mens that players performing well in that period after YE finals might have a better start in th eforum competition.

For example: Gabi Taylor's wonderful $25K win that came in December is still going to be eligible to count in the final reckonging of the 2018 forum competition. She won't have to win another $25K in 2018 to have at least one score of that value count.

On the other hand, Katie Swan's $25K win points from earlier in the year will fall off before the 2018 forum competition runs it course - if she is to have a points score that large to count, she will need to win one again at some point in 2018.

Players could perform sufficiently well in 2018 to supercede all of their scores up to and including any hauls they made in the period after YE finals, and then it wouldn't be a consideration. With Gabi having made that 50-pointer, she would need 16 scores in advance of that. I'd love to see that! It's probably not likely quite yet though.

indiana · Tennis legend

With the WR prediction scores been based on % deviation rather than number of places deviation ( absolutely rightly in my view ) it does mean you have to get very close with hopefully very well ranked players such as Andy and Jo. And clearly in general to score decent points you need to get closer in ranking places the better the player is ranked.

If the player finishes WR 1, 2 or 3 you have to be spot on to get any points. It's a 5 or 0 deal. If they say finish WR 9 even one place out is more than 10%, two places out more than 20% and 3 places out no score.

Bolt · Lower Club Player

J KONTA 13

H WATSON 95

L ROBSON 103

N BROADY 124

K BOULTER 170

K DUNNE 253

K SWAN 256

G TAYLOR 283

M LUMSDEN 302

E SILVA 338

scottie1 · Futures level

KONTA 10

WATSON 65

BROADY 120

BOULTER 143

TAYLOR 160

SWAN 175

ROBSON 201

MOORE 205

DUNNE 210

BURRIDGE 232

blob · All-time great

In light of indiana's valid critique, here is a different way to look at the forums perceived ranking increases for our players in 2018 that shows the number of selections for each player.

The size (area) of the bubble represents the number of selectors for that player as a poportion of the total number of available selectors - now 37 entrants (up to and including scottie1)
This reflects, not the confidence of the perceived 2018 performance, because confidence has a specific statistical meaning, but you can think of it as the collective authority of the forum in the prediction - bigger bubble, greater authority, and vice versa.

We still think everyone will improve.

Tennpar · Strong Club Player

Jo 8
Heather 90
Naomi 111
Gabi 178
Katie S 180
Katie B 245
Jodie 282
Laura 310
Maia 315
Freya 449

Peter too · Tennis legend

Michael D wrote:
Yes for Laura for instance, there have been a wide range of projections for her within the top 10 as you show, ranging from those who think she will improve to those who thing she will decline.... and then there are a further 6 people who think she will fall out of the top 10 altogether, and thus have her projected at 300+, which I'm sure would shift her average overall prediction to one of being a net decliner rather than improver. Similarly and even more so for Tara. It would actually be really interesting to have the 'true' averages of all those who received at least one top ten vote. I gave everyone within the present top 25 predicted rankings on my own list in order to arrive at the top 10.

I could see a whole cluster of players competing for tenth place around a ranking of 350, Tara, Jodie, Maia and Francesca but eventually opted for Sam as the most likely to take that spot.

Nix · Futures level

Here goes!

1. Jo 7
2. Heather 72
3. Naomi 125
4. Katie B 135
5. Gabi Taylor 155
6. Katie Swan 180
7. Laura Robson 195
8. Katy Dunne 210
9.Tara Moore 225
10. Harriet Dart 240

Sim · County player

Predictions table is up and running here: http://www.britishtennis.net/predictions/2018%20Female%20Ranking%20predictions.htm

The current rankings are included as "Status Quo" as last entry.

A very high proportion have gone for the current top 9 with the battle for the final spot pretty evenly split between Tara, Sam and Sarah Beth with Eden not far behind.

EWS is the highest ranked player not to get any picks at GB16

Current standings:

Points Table

Peter too	70
Sim	68
flamingowings	65
cecil	64
Helen40	64
TheTraveller2	63
christ	60
Jon	60
kundalini	60
paulsi	60
brendan	59
Coup Droit	59
indiana	59
Grachka	57
Strongbow	57
SuperT	56
Bolt	55
jb288	55
Nix	55
Brittak	54
Mervatron2	54
JonH	53
LordBrownof	52
cakj1234	51
the addict	51
Canda	50
Jiwan	50
Miriambee	50
theemptyvessel	50
scottie1	49
hoots	47
Tennpar	47
wimdledont	46
Long time reader	45
Michael D	44
Andy Parker	41
telstar	41
vohor	37
Ian1314	34

Sim · County player

Blob has already given lots of impressive stats but here is no. of picks, high and low pick, average and change from current ranking.

As has been pointed out the lower number of picks the less realistic the average is.

All players except Jo-Ko predicted to improve....

	Picks	Total	High	Low	Average	Actual	+/-

Jo Konta	39	391	3	19	10	9	-1
Heather Watson	39	2733	30	130	70	75	5
Naomi Broady	39	4385	60	181	112	116	4
Katie Boulter	39	6118	88	288	157	199	42
Laura Robson	33	6250	75	310	189	228	39
Katy Dunne	38	8862	173	310	233	278	45
Gabriella Taylor	39	8487	125	350	218	284	66
Katie Swan	39	8174	129	265	210	301	91
Harriet Dart	31	8315	131	385	268	307	39
Tara Moore	10	2415	175	300	242	350	109
Samantha Murray	11	3179	199	350	289	424	135
Amanda Carreras	2	555	250	305	278	444	167
Eden Silva	9	3062	300	400	340	501	161
Maia Lumsden	3	992	302	375	331	525	194
Jodie Burrage	5	1564	232	360	313	540	227
Freya Christie	12	3721	249	449	310	547	237
Sarah Beth Grey	1	320	320	320	320	597	277

blob · All-time great

Sim = MVP

Thanks for runnning the competitions again

SuperT · Futures qualifying

blob wrote:
SuperT wrote:
Just out of interest how do you go about deciding who has won? Is it the person who gets the correct order 1 to 10 or the person who most closely matches the year end rankings of individual players? I can't imagine how you'd calculate the latter but since there are clearly some wizard statisticians on this forum I'm sure you'd find a way. It's not that I'm competitive or anything - just as a newbie I'm interested to understand how things work.

Wishing a very happy New Year to all
It's sim's system, and all credit and thanks goes to them.
You get points for how close you get in predicting the GB ranking spot for a player, and a separate set of points for how close you get to predicting the final WTA rank for a player, irrespective of wherther you get the GB ranking place for them correct or not. You can score in one category for a player whilst having got the other completely wrong.
Here, from queries made last year is sim's explanation:
Scoring system:
GB ranking. Spot on = 5pts, 1 out = 4 pts, 2 out = 3 pts, 3 out = 2 pts, 4 out or in top 10 = 1 pt
World ranking. Within 5% = 5pts, 5-10% out = 4 pts, 10-15% out = 3 pts, 15-20% = 2pts, 20-25% = 1pt.
The 1pt for being in GB top 10 was added as the the main aim is to predict who you think will be in the GB top 10.
Note it is possible to get points if a player finishes in GB 11,12,13 or 14 (e.g. you predict 10 and they finish 14 = 1pt.
Updates are usually after each Slam, and the final result after the respective YE Finals.
Here's the 2017 competition page on steven's site: http://www.britishtennis.net/predictions/2017%20Female%20Ranking%20predictions.htm

Thanks for the explanation Blob and thanks to all those who do the work required for the calculations. Think I'll just sit back now and enjoy the ride - and the tennis!

indiana · Tennis legend

Katie Swan WR 301 (CH 296) is the only player man or woman selected by everyone and with a lowest prediction (265) better than their CH.

Jay Clarke WR 255 (CH 219) is the only other player selected by everyone and with a lowest prediction (220) better than their current ranking. In his case that lowest prediction is one place short of his CH.

indiana · Tennis legend

Out of interest, I calculated an average from everyone for Laura and Tara if all the people that didn't put them in their top 10 ( 6 for Laura and 29 for Tara ) ranked them just one WR place below their #10 selection.

Results :

Laura overall then averages 210, as against her average from her 33 selectors of 189 and current ranking of WR 228.

Tara overall then averages 306, as against her average from her 10 selectors of 242 and current ranking 350.

So on that very conservative basis I can't prove that either's 'true' overall average would be lower than their current ranking. But certainly more towards it than my figures and much more towards it than the average from their selectors.