Thursday, August 11, 2016

The Electoral College Map (8/11/16)




New State Polls (8/11/16)
State
Poll
Date
Margin of Error
Sample
Clinton
Trump
Undecided
Poll Margin
FHQ Margin
Alabama
6/19
+/- 2.0%
4100 adults
33
57
--
+24
+24.00
Florida
8/10
+/- 4.0%
622 likely voters
45
44
3
+1
+2.19
Georgia
8/4-8/8
+/- 2.5%
1604 likely voters
39
43
--
+4
+0.60
Idaho
5/18-6/4
+/- 3.99%
603 adults
32
49
18
+17
+19.36
Iowa
8/8-8/10
+/- 4.4%
500 likely voters
36
37
18
+1
+2.04
Maine
8/4-8/8
+/- 2.2%
2046 likely voters
43
33
--
+10
+8.69
New Hampshire
8/7-8/8
+/- 3.4%
820 registered voters
41
31
13
+10
+4.60
New York
8/4-8/8
+/- 2.4%
1717 likely voters
48
34
--
+14
+19.17
South Carolina
8/9-8/10
+/- 2.7%
1290 likely voters
39
41
13
+2
+2.00


Polling Quick Hits:
There was a flood of new polling from old and new battleground states alike and a couple of backlogged polls freshly added to the FHQ dataset. Here is a quick tour.


Alabama:
One of the older polls comes out Dixie. There is quite a bit of support not going to the two major party candidates, but Trump led Clinton by 24 points. Both are underperforming where their 2012 predecessors were in the Yellowhammer state, but the margin is in line with the spread in November 2012. Even if there is a blue wave that surges into the most Republican columns below on the Electoral College Spectrum, Alabama figures to be one the hold outs on the Republican side of the partisan line.


Florida:
One has to wonder if the volatility in the polling in Florida is dying down some after the narrow margins of the last two days of survey releases. That said, Quinnipiac -- yesterday's entrant -- has leaned more Republican this cycle. It could be that there is a tighter new normal in the Sunshine state, or that could be a function of the firms that have released polls in the last two days. It will just take some more data to answer that question.


Georgia:
In the Peach state, there is yet more evidence that Georgia is more competitive in 2016. The head-to-head in the new Gravis poll has Trump clinging to a one point advantage, but the multicandidate results show a larger gap with most of the support for the third party candidates being siphoned off of Clinton's. Either way one goes -- and FHQ uses the multicandidate numbers in these situations -- Trump is just a hair ahead of Clinton in the FHQ averages. Georgia is essentially tied along with Arizona and Nevada.


Idaho:
Dan Jones and Associates polled Idaho just before the end of primary season. As was the story in the firm's July survey of the Gem state, Trump was ahead by a comfortable margin, but running behind recent GOP nominees in the state. Like Alabama above, Idaho will likely be red in the fall, insulated from any Clinton advances. It is only August, but these are data from before the conventions.


Iowa:
Like Florida, another day brought another close poll in the Hawkeye state. The slight Trump edge nudged the FHQ average down enough that Florida and Iowa switched places with each other on the Spectrum, and both are now fairly tightly knotted in a group with Ohio in terms of their average margins. What is perhaps more interesting about Iowa is that no candidate has garnered anything more than 44% in any poll this year (with the exception of the very first survey back in early January). And now both candidates in the last two days of polling have dipped into the 30s. Iowa is close, but there is a deep reservoir of undecideds there. That should taper off as election day approaches, but could swing Iowa depending on how that segment breaks.


Maine:
There has not been a lot of survey work done in Maine yet. Thus far in 2016, there has been but one poll, and it showed Clinton ahead, though in the lean range. Another poll -- this one from Gravis -- has the race in the Pine Tree state in a similar position, but with a slightly larger Clinton lead. That was enough to bump the FHQ average in the state up, but keep Maine in the Lean Clinton category.


New Hampshire:
If Florida and Iowa have had a series of closer post-convention polls, New Hampshire can be said to be moving in the opposite direction. Rather than a second close poll, there is now a second double digit lead for Clinton in the Granite state in the wake of the two national conventions. One the weight of those polls, New Hampshire enters the Watch List -- on the verge of shifting into a Lean Clinton state -- and pushes past Virginia on the Electoral College Spectrum, breaking the electoral college "tie" scenario that occurs when Pennsylvania and Virginia are on the Democratic side of New Hampshire. [Tie is in quotation marks above because the partisan line pushes well beyond where those three states are in the alignment of states. It is not at all a likely scenario at this point in time.]


New York:
There are a number of ways one could frame the current picture in New York: home state of both candidates, tighter than usual margin, etc. But it is pretty clear that Trump is hovering right around where Romney did in the Empire state in 2012. That makes New York a story of Clinton underperforming Obama in the state. Yet, even given that trend, the margin there is comfortably blue. It would be more noteworthy if Trump was actually running ahead of Romney's pace there. He isn't and that is likely to keep New York in the Democratic column.


South Carolina:
Changes (August 11)
StateBeforeAfter
South CarolinaLean TrumpToss Up Trump
Finally, the first presidential polling data of 2016 emerged from South Carolina. It might be easy to get carried away with the small two point advantage Trump has in the Palmetto state. And it is a surprising number at first glance. However, South Carolina ended up a couple of percentage points redder than Georgia in 2012. And if Georgia is tied, then South Carolina +2 to Trump is right in line with what would be a uniform shift of states. And truth be told, that is not that far off. South Carolina is still in range of its positioning on the Spectrum in November 2012. It is simply closer there in 2016 if much stock can be put in just one survey. Of course, this one does harken back to the 2008 Zogby poll of South Carolina; the one where Obama was narrowly ahead.

South Carolina shifts from Lean to Toss Up Trump with the addition of this poll, and jumps to a spot next to Georgia on the Spectrum after the Peach state switched places with Arizona.





The Electoral College Spectrum1
HI-42
(7)
NJ-14
(175)
NH-4
(260)
MS-6
(155)
TN-11
(58)
MD-10
(17)
DE-3
(178)
VA-133
(273 | 278)
MO-10
(149)
LA-8
(47)
RI-4
(21)
WI-10
(188)
FL-29
(302 | 265)
AK-3
(139)
SD-3
(39)
MA-11
(32)
ME-4
(192)
IA-6
(308 | 236)
UT-6
(136)
ND-3
(36)
VT-3
(35)
NM-5
(197)
OH-18
(326 | 230)
KS-6
(130)
ID-4
(33)
CA-55
(90)
MI-16
(213)
NC-15
(341 | 212)
IN-11
(124)
NE-5
(29)
NY-29
(119)
OR-7
(220)
NV-6
(197)
TX-38
(113)
AL-9
(24)
IL-20
(139)
CT-7
(227)
AZ-11
(191)
AR-6
(75)
OK-7
(15)
MN-10
(149)
CO-9
(236)
GA-16
(180)
MT-3
(69)
WV-5
(8)
WA-12
(161)
PA-20
(256)
SC-9
(164)
KY-8
(66)
WY-3
(3)
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he or she won all the states ranked prior to that state. If, for example, Trump won all the states up to and including Virginia (all Clinton's toss up states plus Virginia), he would have 278 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Clinton's number is on the left and Trumps's is on the right in bold italics.


To keep the figure to 50 cells, Washington, DC and its three electoral votes are included in the beginning total on the Democratic side of the spectrum. The District has historically been the most Democratic state in the Electoral College.

3 Virgini
a is the state where Clinton crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line.



NOTE: Distinctions are made between states based on how much they favor one candidate or another. States with a margin greater than 10 percent between Clinton and Trump are "Strong" states. Those with a margin of 5 to 10 percent "Lean" toward one of the two (presumptive) nominees. Finally, states with a spread in the graduated weighted averages of both the candidates' shares of polling support less than 5 percent are "Toss Up" states. The darker a state is shaded in any of the figures here, the more strongly it is aligned with one of the candidates. Not all states along or near the boundaries between categories are close to pushing over into a neighboring group. Those most likely to switch -- those within a percentage point of the various lines of demarcation -- are included on the Watch List below.


The Watch List1
State
Switch
Alaska
from Lean Trump
to Toss Up Trump
Arizona
from Toss Up Trump
to Toss Up Clinton
Arkansas
from Strong Trump
to Lean Trump
Georgia
from Toss Up Trump
to Toss Up Clinton
Missouri
from Toss Up Trump
to Lean Trump
Nevada
from Toss Up Trump
to Toss Up Clinton
New Hampshire
from Toss Up Clinton
to Lean Clinton
New Jersey
from Strong Clinton
to Lean Clinton
Virginia
from Toss Up Clinton
to Lean Clinton
Wisconsin
from Lean Clinton
to Strong Clinton
1 Graduated weighted average margin within a fraction of a point of changing categories.





Wednesday, August 10, 2016

The Electoral College Map (8/10/16)




New State Polls (8/10/16)
State
Poll
Date
Margin of Error
Sample
Clinton
Trump
Undecided
Poll Margin
FHQ Margin
Iowa
8/3-8/7
+/- 3.1%
899 likely voters
35
35
12
+0
+2.48
Kansas
8/3-8/7
+/- 4.2%
566 likely voters
39
44
7
+5
+7.38
Ohio
8/3-8/7
+/- 3.1%
889 likely voters
39
35
10
+4
+2.02
Pennsylvania
8/3-8/7
+/- 3.1%
834 likely voters
45
36
7
+9
+6.16
Wisconsin
8/4-8/7
+/- 4.6%
683 likely voters
47
34
6
+13
+9.56


Polling Quick Hits:
A couple of battleground state polls, a handful of leaners and another set of reinforcing polls.

Iowa:
The Hawkeye state continues to be underpolled, but other than one outlier that bumped the FHQ average up, Iowa has resided closer to the partisan line and on the Republican side of states like Colorado, Pennsylvania and Virginia in 2016 than it did four years ago. It has been blue throughout but more competitive. Iowa, then, is more like a state such as Nevada, where Clinton's support has been lower relative to Obama's in 2012. That is borne out in this latest poll from Marist that shows a tie in a multiway race. Still, Iowa is nearly two and a half points on the Clinton side of the partisan line on the Electoral College Spectrum.


Kansas:
Just when one thought Kansas was reverting to a more entrenched Republican state, another poll emerges showing a narrow Trump lead in the state. In fact, since Survey USA was last in the field in the Sunflower state just before the conventions, Trump's advantage has contracted, swinging six points toward Clinton. The New York businessman continues to lead, but if he is ahead by only five in Kansas -- and assuming the states are lined up in the order they are below -- then Clinton blue is more likely to stretch further in to that middle column and claim another few states.

Yes, statewide polling in Kansas in 2014 -- in both the gubernatorial and senate races -- was competitive before breaking Republican in the end. This may, then, just be a Kansas thing. However, it could also signal that additional polling is warranted in some of the other traditionally ruby red states.


Ohio:
In the 12 surveys conducted in the Buckeye state since the end of primary season in early June, Clinton has led in eight and been tied in another one. The new Marist poll of Ohio is one of those eight and buttresses the notion that Clinton leads somewhere in the range of tied to five points ahead. The FHQ graduated weighted average and this poll sit right in that area.

Additionally, it should be said that this poll was enough to push Ohio back on the Democratic side of North Carolina in the alignment of states; something that has tended to be true in the previous cycles of most of the recent past.


Pennsylvania:
Like Ohio, the Marist poll of the Keystone state does nothing to shake the premise that Clinton's lead has expanded in the state since the conventions. This is another poll with the former secretary of state up by close to ten points that further solidifies Pennsylvania's position within the Lean Clinton category. And it is enough to push the commonwealth off the Watch List. No longer is Pennsylvania within range of shifting back into the toss up area. Further data may change that outlook, but for the time being, Pennsylvania -- a valuable 20 electoral votes in the path to 270 -- seems even more out of reach for Trump and the Republicans.


Wisconsin:
As was the case in Kansas, the new Marquette poll of Wisconsin witnessed a middle range (seven point) shift toward Clinton since its pre-conventions poll. Unlike Kansas, however, polling already had the Badger state firmly within the Clinton group of states. But that seven point shift since the last Marquette survey helps to push the average back to within range of crossing over into the Strong Clinton category. The other states in the Lean Clinton category are all comfortably within its bounds, but Wisconsin now replaces Pennsylvania as the only lean state on the Clinton side of the partisan line close to switching. Only, Wisconsin is not anywhere close to becoming a toss up as Pennsylvania had been. Additional data may change that, but for now, Wisconsin is not looking at all like a swing state despite often being lumped in with that roughly Rust Belt group of states the Trump campaign is said to be targeting.






The Electoral College Spectrum1
HI-42
(7)
NJ-14
(175)
VA-133
(269 | 282)
MO-10
(158)
TN-11
(58)
MD-10
(17)
DE-3
(178)
NH-43
(273 | 269)
AK-3
(148)
LA-8
(47)
RI-4
(21)
WI-10
(188)
IA-6
(279 | 265)
UT-6
(145)
SD-3
(39)
MA-11
(32)
NM-5
(193)
FL-29
(308 | 259)
KS-6
(139)
ND-3
(36)
VT-3
(35)
MI-16
(209)
OH-18
(326 | 230)
IN-11
(133)
ID-4
(33)
CA-55
(90)
OR-7
(216)
NC-15
(341 | 212)
TX-38
(122)
NE-5
(29)
NY-29
(119)
CT-7
(223)
NV-6
(197)
SC-9
(84)
AL-9
(24)
IL-20
(139)
ME-4
(227)
GA-16
(191)
AR-6
(75)
OK-7
(15)
MN-10
(149)
CO-9
(236)
AZ-11
(175)
MT-3
(69)
WV-5
(8)
WA-12
(161)
PA-20
(256)
MS-6
(164)
KY-8
(66)
WY-3
(3)
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he or she won all the states ranked prior to that state. If, for example, Trump won all the states up to and including Virginia (all Clinton's toss up states plus Virginia), he would have 282 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Clinton's number is on the left and Trumps's is on the right in bold italics.


To keep the figure to 50 cells, Washington, DC and its three electoral votes are included in the beginning total on the Democratic side of the spectrum. The District has historically been the most Democratic state in the Electoral College.

3 New Hampshire and Virginia are collectively the states where Clinton crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line. If those two states are separated with Clinton winning Virginia and Trump, New Hampshire, then there would be a tie in the Electoral College.



NOTE: Distinctions are made between states based on how much they favor one candidate or another. States with a margin greater than 10 percent between Clinton and Trump are "Strong" states. Those with a margin of 5 to 10 percent "Lean" toward one of the two (presumptive) nominees. Finally, states with a spread in the graduated weighted averages of both the candidates' shares of polling support less than 5 percent are "Toss Up" states. The darker a state is shaded in any of the figures here, the more strongly it is aligned with one of the candidates. Not all states along or near the boundaries between categories are close to pushing over into a neighboring group. Those most likely to switch -- those within a percentage point of the various lines of demarcation -- are included on the Watch List below.


The Watch List1
State
Switch
Alaska
from Lean Trump
to Toss Up Trump
Arizona
from Toss Up Trump
to Toss Up Clinton
Arkansas
from Strong Trump
to Lean Trump
Georgia
from Toss Up Trump
to Toss Up Clinton
Missouri
from Toss Up Trump
to Lean Trump
Nevada
from Toss Up Trump
to Toss Up Clinton
New Jersey
from Strong Clinton
to Lean Clinton
Virginia
from Toss Up Clinton
to Lean Clinton
Wisconsin
from Lean Clinton
to Strong Clinton
1 Graduated weighted average margin within a fraction of a point of changing categories.





Tuesday, August 9, 2016

The Electoral College Map (8/9/16)




New State Polls (8/9/16)
State
Poll
Date
Margin of Error
Sample
Clinton
Trump
Undecided
Poll Margin
FHQ Margin
Florida
7/30-8/7
+/- 3.0%
1056 likely voters
43
43
3
+0
+2.27
Illinois
7/11-7/14
+/- 3.5%
800 likely voters
51
32
--
+18
--
Illinois
8/1-8/4
+/- 3.5%
800 likely voters
51
32
--
+18
+19.03
Kentucky
7/31-8/1
+/- 4.4%
500 registered voters
36
49
7
+13
+13.00
Missouri
8/5-8/6
+/- 3.0%
1280 likely voters
42
44
7
+2
+4.50
North Carolina
8/5-8/7
+/- 3.4%
830 likely voters
43
41
8
+2
+1.88
Ohio
7/30-8/7
+/- 3.4%
812 likely voters
44
42
4
+2
+1.87
Pennsylvania
7/30-8/7
+/- 3.4%
815 likely voters
48
39
3
+9
+5.95


Polling Quick Hits:
Eight new polls from seven states are new to the FHQ polling dataset. The bulk of them are reinforcing polls; those within range of the FHQ graduated weighted average margin in the states.


Florida:
Three of the those polls -- battleground state surveys from Quinnipiac -- fell mostly in line with where the majority of post-convention polls have found the three most electoral vote-rich states. Florida and Ohio are tightly contested and Pennsylvania appears to be reverting to form as the margin in the Keystone state continues to grow for Clinton.

Though Florida and Ohio are close, Florida has tended to lean a bit more heavily toward Clinton given the full universe of 2016 data from both states. That is not the case here as Quinnipiac finds Florida tied and Clinton with a small lead in Ohio. But then the previous Quinnipiac surveys have had a seemingly Republican-leaning house effect relative to other polling outlets. The transposing of Florida and Ohio may be a function of that. Of course, it should be noted that Florida swung more intensely toward Clinton since the last Q-poll in the Sunshine state (+5 in Clinton's direction) than in Ohio (+3 toward Clinton since July).


Illinois:
There is very little to say about the two newly added polls out of the Land of Lincoln. Both from Democratic firm, Norrington Petts (one from July and one from August), find Clinton ahead by a similar margin to Illinois survey a month ago. Additionally, both polls show a race stuck in neutral in that time with Clinton up 19. Both nominees are running behind their 2012 counterparts, but the margin is approximately where it was in November 2012: comfortably Democratic.


Kentucky:
The first look at neighboring Kentucky indicates a similarly comfortable, albeit, Republican state. Harper's survey has Trump ahead by double digits, but has the New York businessman lagging about ten points behind the pace Romney established in 2012. Clinton, meanwhile, is in range of Obama 2012. Under normal circumstances one might expect any undecideds to break toward the Republican, but that is less than clear in a cycle that has proven anything but normal thus far.


Missouri:
Changes (August 9)
StateBeforeAfter
MissouriLean TrumpToss Up Trump
Another poll in from the Show-Me state makes that recent Trump +10 from Survey USA seem more and more out of step. The latest from Remington finds Trump ahead by a narrower two points and much closer to the July Mason-Dixon poll that had Clinton ahead by one. Missouri's a state that had moved toward the Republicans over the last few cycles and away from the quadrennial bellwether it had once been. Still, like a number of states in its general vicinity on the Electoral College Spectrum below (see especially Georgia), the average in Missouri has contracted some in 2016. Rather than being firmly planted in the Lean Republican area, the Show-Me state has instead jumped back and forth across the toss up/lean line on the Republican side. Today's poll was enough to push it back over into the Toss Up Trump category, but keep it on the Watch List (within a fraction of a point of moving back into Lean Trump territory).


North Carolina:
PPP has made up a little more than a third of the 2016 polling in North Carolina, and while other firms' polls have jumped around relative to each other, the PPP surveys have always seen a close race between Clinton and Trump in the Tar Heel state. This August poll is no exception. The difference is that Clinton is now the candidate with the lead, and one in this poll that is right on the FHQ average in the state: close but on the Clinton side of the partisan line.


Ohio:
And the interesting thing now is that Ohio is now right there with North Carolina after a period in which there had been some space between the two in the FHQ averages. Like the PPP North Carolina poll, the Q-poll of Ohio is right in line with where the average is in the state. And the one trend that is clear here and in all the states above is that the polling is tilting in the immediate aftermath of the conventions and Trump's rough week last week toward Clinton.


Pennsylvania:
Another day and another poll out of Pennsylvania showing Clinton up by a margin hovering around the strong-lean line (+10 points). But this one represents quite a shift, in-house. The last Quinnipiac survey of the commonwealth showed Trump up six points in a multi-candidate race. However, this month Clinton is ahead nine; a 15 point shift. Although it has moved in the same direction as the Florida poll -- toward Clinton -- the change is three times greater. That is a pretty significant change month over month. One that for the first time brings the Quinnipiac numbers much more in line with the rest of the current polling in the Keystone state. Again, Pennsylvania is a state that Trump needs to get to 270. And if Quinnipiac is showing a nearly 10 point advantage for Clinton, well...

Pennsylvania retains its position on the spectrum below and inches closer still to moving off the Watch List into a safer -- from the Clinton perspective -- position within the Lean category.





The Electoral College Spectrum1
HI-42
(7)
NJ-14
(175)
VA-133
(269 | 282)
MO-10
(158)
TN-11
(58)
MD-10
(17)
DE-3
(178)
NH-43
(273 | 269)
AK-3
(148)
LA-8
(47)
RI-4
(21)
WI-10
(188)
IA-6
(279 | 265)
UT-6
(145)
SD-3
(39)
MA-11
(32)
NM-5
(193)
FL-29
(308 | 259)
IN-11
(139)
ND-3
(36)
VT-3
(35)
MI-16
(209)
NC-15
(323 | 230)
TX-38
(128)
ID-4
(33)
CA-55
(90)
OR-7
(216)
OH-18
(341 | 215)
KS-6
(90)
NE-5
(29)
NY-29
(119)
CT-7
(223)
NV-6
(197)
SC-9
(84)
AL-9
(24)
IL-20
(139)
ME-4
(227)
GA-16
(191)
AR-6
(75)
OK-7
(15)
MN-10
(149)
CO-9
(236)
AZ-11
(175)
MT-3
(69)
WV-5
(8)
WA-12
(161)
PA-20
(256)
MS-6
(164)
KY-8
(66)
WY-3
(3)
1 Follow the link for a detailed explanation on how to read the Electoral College Spectrum.

2 The numbers in the parentheses refer to the number of electoral votes a candidate would have if he or she won all the states ranked prior to that state. If, for example, Trump won all the states up to and including Virginia (all Clinton's toss up states plus Virginia), he would have 282 electoral votes. Trump's numbers are only totaled through the states he would need in order to get to 270. In those cases, Clinton's number is on the left and Trumps's is on the right in bold italics.


To keep the figure to 50 cells, Washington, DC and its three electoral votes are included in the beginning total on the Democratic side of the spectrum. The District has historically been the most Democratic state in the Electoral College.

3 New Hampshire and Virginia are collectively the states where Clinton crosses the 270 electoral vote threshold to win the presidential election. That line is referred to as the victory line. If those two states are separated with Clinton winning Virginia and Trump, New Hampshire, then there would be a tie in the Electoral College.



NOTE: Distinctions are made between states based on how much they favor one candidate or another. States with a margin greater than 10 percent between Clinton and Trump are "Strong" states. Those with a margin of 5 to 10 percent "Lean" toward one of the two (presumptive) nominees. Finally, states with a spread in the graduated weighted averages of both the candidates' shares of polling support less than 5 percent are "Toss Up" states. The darker a state is shaded in any of the figures here, the more strongly it is aligned with one of the candidates. Not all states along or near the boundaries between categories are close to pushing over into a neighboring group. Those most likely to switch -- those within a percentage point of the various lines of demarcation -- are included on the Watch List below.


*Due to the way in which states with no polling are treated in 2016 by FHQ -- uniform swing -- South Carolina has seen its "average" margin shrink. In the last month, as more polling data has been accrued, that average uniform swing has increased to just a shade more than one and a half points toward Clinton. That development has pushed South Carolina further away from the Strong Trump category and off the Watch List below.


The Watch List1
State
Switch
Alaska
from Lean Trump
to Toss Up Trump
Arizona
from Toss Up Trump
to Toss Up Clinton
Arkansas
from Strong Trump
to Lean Trump
Georgia
from Toss Up Trump
to Toss Up Clinton
Missouri
from Toss Up Trump
to Lean Trump
Nevada
from Toss Up Trump
to Toss Up Clinton
New Jersey
from Strong Clinton
to Lean Clinton
Pennsylvania
from Lean Clinton
to Toss Up Clinton
Virginia
from Toss Up Clinton
to Lean Clinton
1 Graduated weighted average margin within a fraction of a point of changing categories.