From

Location

Message

Date Posted

South Hill, Virginia

"The Chart", though surprisingly accurate, is a result of weights and measures of pumpkins...not the other way around. My guess is that some sort of regression is done to come up with the chart and as data is added the chart is updated. Typically with regression the more data you have the more accurate it will be.

AGs come in many shapes and this makes estimating weight difficult when you use the same measuring process for all of them. A pumpkin that weighs significantly over or under the chart may just be an odd shape for its weight or there may not be significant data for pumpkins of that size.

Since the chart changes as more weights are recorded (as it should) a pumpkin that was formerly over the charts could end up under the charts (or the other way around). The weights of many of the “EST” designated fruit will change too. Suppose that a pumpkin that weighed 10% under one year weighed 10% over the next year because of adjustments to the chart. Does that seed somehow become better?

I’m not trying to dismiss wall thickness and high density as desirable factors in a big pumpkin but the chart is not necessarily going to be any indication of either. I’m not even sure they are genetic traits but I’ll have to leave that topic to the resident scientists we have around here. Thick walls are nice but if you don’t get “big” too then who cares. Would you rather have a seed that produced a 1200 pound “under chart” pumpkin or a seed that produced a 100 pound “over chart” pumpkin?

8/25/2003 5:34:58 PM

Ga.

Brent,,just my two cents worth..When a pkin goes light,,is it because the fruit wasnt picked in its prime??Because it was left on vine in the heat too long after its prime?? I ask this because 600 pistono went light,,but its offspring went a little heavy..So just because a fruit goes light,,doesnt mean its offspring will also...
J

8/26/2003 7:32:50 AM

Utah

after a while more data points will not give you more accuracy. variation is there and more points won't make it go away. The number of points required to get excellent equations is surprizingly small. like between 50 and 100. after that more data points don't really help- except at the high end of the scale, where we keep getting new points that we must take into account for any new equation. two new really heavy fruit that are either both heavy or both light could cause the new equation to change alot. a change that would not be acurate... and we wouldn't know till we obtained more fruit of that same size that weigh where they should.
mean while 500 or 1000 new fruit in the middle won't change the equations at all ... UNLESS...Our fruit are collectively getting heavier.

all this leads to other questions.... if our fruit are all getting heavier then should a new equation only include the data the recent data... say the last two or three years?
or is it just a few of the new really heavy fruit that is causing the equations to change? and is that really bad change?

8/26/2003 8:52:53 AM

Utah

after a while more data points will not give you more accuracy. variation is there and more points won't make it go away. The number of points required to get excellent equations is surprizingly small. like between 50 and 100. after that more data points don't really help- except at the high end of the scale, where we keep getting new points that we must take into account for any new equation. two new really heavy fruit that are either both heavy or both light could cause the new equation to change alot. a change that would not be acurate... and we wouldn't know till we obtained more fruit of that same size that weigh where they should.
mean while 500 or 1000 new fruit in the middle won't change the equations at all ... UNLESS...Our fruit are collectively getting heavier.

all this leads to other questions.... if our fruit are all getting heavier then should a new equation only include the data the recent data... say the last two or three years?
or is it just a few of the new really heavy fruit that is causing the equations to change? and is that really a bad change?

8/26/2003 8:53:31 AM

Deer Park, WI

Interesting question Brent- Here's my take. I'd rather have a seed that CONSISTENTLY produced 100 lb "over chart" pumpkins, rather than one that consistently produced 1200 lb "under chart" pumpkins. The fact that a seed can produce multiple 1200lbers is a combintation of genetics and awesome environment. We already have many of those seeds in circulation. However, finding a seed that can repeatedly produce 100lb heavy fruit is something special, and has not been done yet. Whether % heavy is a function of genetics or enviroment, we do not know. As with most things, it seems to be a mixture of both. There are some out there (730 Stell, 805 Pukos) that in general, tend to go heavy. Of course, there are also those (801 Stelts) that tend to go light. However, we have yet to find a seed (to my knowledge) that produces fruit going consistently heavy (>10%) 100% of the time. I'd trade my collection for one of these.

Regarding the charts, it would be possible to have a set of data for each fruit shape. Say we have 3 basic shapes, barrels, wheels, and low riders. There could be criteria that would qualify a fruit for the shape, based on its OTT measurments. For example, in order to qualify as a barrel, the Front to back measurment has to be "X" inches greater than the side to side. This type of system would be much more precise than the one-size-fits-all that we have now, but would require a tremendous amount work to develop.

8/26/2003 9:05:56 AM

Garden State Pumpkins

Good Point Joe...but on the other hand if everyone grew those seeds eventually the chart would even out and they wouldn't be 10% heavy anymore ...just normal..I'll betcha the old timers out there with all the experience can guess a weight by eyeballing size and shape and knocking on it for sound without a chart...Something I'll never be able to do, they all look huge to me...so the chart is my best guesstimate as to where they stand...I think as long as they're grown the true weight will be at the Scale.

8/26/2003 10:59:53 AM

Cooks Creek, Manitoba, Canada

I'm assuming that the chart created by Len Stellpflug was done using some form of regression analysis. Does anyone know what type was used and what the R squared value is?

8/26/2003 11:09:15 AM

Caceres, mid west of Spain

I believe the chart to be very well done and work in a very accurated way, statistacally, of course. Because a particular fruit can go until 23% heavier or ligther than the estimated (as long as we know so far). Then, what we need is to know if our particular fruit will be heavy or light. Experienced growers with many fruits in their backs can guess it much better than newbies, because some tips like skin, shape, shoulders... give you an idea. But it would be nice to have a test that inform you about the wall thickness and density, and that shouldn't be very complicated, because there are ustrasonic apparatus (I used them once to test tree stability, measuring the inner hole width) that can do that. They aren't cheap though.

8/26/2003 12:10:21 PM

Garden State Pumpkins

Isn't the chart just to give you an idea of how many cases of beer you need to have so there'e enough for all the guys that come help you get it out of the garden to go weigh it!!

8/26/2003 3:31:44 PM

Columbus, Ohio

We finally hit the nail on the head with that answer...the chart is merely used to indicate appropriate alcohol consumption! Bravo!

8/26/2003 5:03:11 PM

Liverpool , England

If the chart was 100% accurate then we would not need to weigh them. Now where's the fun gone.

Alun

8/26/2003 8:40:15 PM

Garden State Pumpkins

Sorry Brent Thread drift!

8/27/2003 7:46:26 AM

Knopp, Germany

Prior to weighing my fruit on the 805 last week, (went 6.3% heavy), I had a good feeling about it going heavy. Now I am not a very experienced grower, but compared to the SS and EE measurements, the Circ was fairly big. The shoulders where also very blocky, not your typical high shouder look. After tapping it on the blossom end, sounded hollow, and then near the top, sounded very solid. I realized that about 75% of the fruit had the solid sound. So, in this case I felt good about guessing it was going to go heavy.

I then went to the AGGC and looked at all of the progeny from the 805 and realized that this fruit has consistently gone heavy. Looking at those pictures that were available on the site also pointed out that it typically grows in the paatern mine had grown, for fruit of its weight.

So, I guess what I am saying is that the charts are probably closer for fruit that grow more typical in shape, and that for those that are blocky or flat, then the chances of the fruit going heavy are higher. I would love to know if crossing the 805 with a fruit that has produced a big pumpkin that went light will give the resulting seeds any heavy tendencies?

Bottom line is I agree, rather a seed that consistently goes heavy, that is why I am making every effort to find another 805 Pukos to grow again next year.

8/27/2003 8:17:41 AM