Addition on the IMO

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

1358 posts

#1 h Sep 22, 2020, 6:28 PM • 31 Y

Y by Smileyklaws, ilovepizza2020, OlympusHero, tree_3, dchenmathcounts, amar_04, A-Thought-Of-God, Shrawon360, Aimingformygoal, Aritra12, phxbesx0705, samrocksnature, Jc426, Yugoo, centslordm, brickmaster8, megarnie, tigerzhang, HWenslawski, son7, rayfish, ImSh95, Lamboreghini, crazyeyemoody907, Mogmog8, aidan0626, deplasmanyollari, tigeryong, Rounak_iitr, Sedro, ItsBesi

Consider the convex quadrilateral $ABCD$ . The point $P$ is in the interior of $ABCD$ . The following ratio equalities hold:
$\angle PAD:\angle PBA:\angle DPA=1:2:3=\angle CBP:\angle BAP:\angle BPC$ Prove that the following three lines meet in a point: the internal bisectors of angles $\angle ADP$ and $\angle PCB$ and the perpendicular bisector of segment $AB$ .

Proposed by Dominik Burek, Poland

This post has been edited 2 times. Last edited by djmathman, Oct 6, 2020, 5:51 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

khina

993 posts

khina

#2 h Sep 22, 2020, 6:30 PM • 4 Y

Y by samrocksnature, Vaddokx, ImSh95, Math_.only.

lol Click to reveal hidden text

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

lrjr24

966 posts

lrjr24

#3 h Sep 22, 2020, 6:33 PM • 4 Y

Y by samrocksnature, Math4Life2020, ImSh95, Nuran2010

Doesn't the announcement say to wait til 8:00 PM EST?
@below ok.

This post has been edited 1 time. Last edited by lrjr24, Sep 22, 2020, 6:34 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

naman12

1358 posts

naman12

#4 h Sep 22, 2020, 6:33 PM • 2 Y

Y by samrocksnature, ImSh95

lrjr24 wrote:

Doesn't the announcement say to wait til 8:00 PM EST?

The problems have been posted on the official website, and thus discussion should be allowed.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

jj_ca888

2726 posts

jj_ca888

#5 h Sep 22, 2020, 6:34 PM • 2 Y

Y by samrocksnature, ImSh95

lrjr24 wrote:

Doesn't the announcement say to wait til 8:00 PM EST?

I actually don't think discussion should be allowed. It's not 23:59 UTC yet.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

timon92

224 posts

timon92

#6 h Sep 22, 2020, 6:37 PM • 12 Y

Y by 62861, Aryan-23, 508669, Kanep, qubatae, samrocksnature, CrazyMathMan, megarnie, ImSh95, aidan0626, Mango247, ehuseyinyigit

This problem was proposed by Burii.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Pluto1708

1107 posts

Pluto1708

#7 h Sep 22, 2020, 6:45 PM • 8 Y

Y by A-Thought-Of-God, samrocksnature, Vaddokx, megarnie, ImSh95, Aryan-23, Mango247, Math_.only.

naman12 wrote:

Consider the convex quadrilateral $ABCD$ . The point $P$ is in the interior of $ABCD$ . The following ratio equalities hold:
$\angle PAD:\angle PBA:\angle DPA=1:2:3=\angle CBP:\angle BAP:\angle BPC$ Prove that the following three lines meet in a point: the internal bisectors of angles $\angle ADP$ and $\angle PCB$ and the perpendicular bisector of segment $AB$ .

Solution

This post has been edited 1 time. Last edited by Pluto1708, Sep 22, 2020, 6:47 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

djmathman

7936 posts

djmathman

#8 h Sep 22, 2020, 6:58 PM • 9 Y

Y by Pluto1708, OlympusHero, karitoshi, samrocksnature, Jc426, HamstPan38825, megarnie, ImSh95, ehuseyinyigit

Definitely European geometry all right. I'll edit in a diagram at some point, but for now:

Choose points $S$ on $\overline{BC}$ and $T$ on $\overline{AD}$ so that $\triangle BSP$ and $\triangle ATP$ are both isosceles. The given angle conditions imply $\angle PSC = \angle SPC$ and $\angle DPT=\angle DTP$ , so the given angle bisectors are actually the perpendicular bisectors of $\overline{PS}$ and $\overline{PT}$ .

But also $S$ and $T$ are the midpoints of the arcs $\widehat{PB}$ and $\widehat{PA}$ of $\odot(PAB)$ , so these bisectors concur at the center of $\odot(PAB)$ , which, by definition, lies on the perpendicular bisector of $\overline{AB}$ .

This post has been edited 1 time. Last edited by djmathman, Sep 22, 2020, 7:01 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

InkretBear

85 posts

InkretBear

#9 h Sep 22, 2020, 7:00 PM • 4 Y

Y by Pluto1708, Steff9, samrocksnature, ImSh95

Let the circumcircle of $PAB$ intersect $AD,BC$ at $X$ and $Y$ respectively. It can be seen by angle chase that $XD=PD, YC=PC$ so the angle bisectors in the problem are just bisectors of $PX$ and $PY$ . Now the three lines concur since they are perpendiculars of chords of a common circle ( $ABYPX$ ).

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

itslumi

284 posts

itslumi

#10 h Sep 22, 2020, 7:08 PM • 7 Y

Y by RudraRockstar, samrocksnature, ImSh95, Complete_quadrilateral, mistakesinsolutions, ihatemath123, pokpokben

What a disaster.

This post has been edited 2 times. Last edited by itslumi, Sep 22, 2020, 9:24 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

mathlogician

1051 posts

mathlogician

#12 h Sep 22, 2020, 7:09 PM • 6 Y

Y by Ya_pank, samrocksnature, Complete_quadrilateral, Mango247, Mango247, Mango247

Truly the pinnacle of olympiad geometry.

Let $O$ be the circumcenter of $(PAB).$ Now using the angle conditions note that $(POAD)$ is cyclic, so an angle chase shows $DO$ bisects $\angle ADP$ . Similarly $CO$ bisects $\angle BCP$ .

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

djmathman

7936 posts

djmathman

#14 h Sep 22, 2020, 7:17 PM • 19 Y

Y by timon92, OlympusHero, Kanep, IAmTheHazard, crazyeyemoody907, myh2910, Ronin_The_Nameless_Ninja, samrocksnature, Jc426, HamstPan38825, 554183, centslordm, tigerzhang, rayfish, CyclicISLscelesTrapezoid, mathmax12, Mango247, aidan0626, OronSH

itslumi wrote:

This isn't for a p1,What a disaster(very easy).

Wait, I actually don't think this is a bad problem (though perhaps not the pinnacle of Euclidean geometry). While the diagram is definitely... artificial, you actually need to understand how to interpret all the angle conditions in order to make any sort of progress on this problem. In my case, I drew all the angle bisectors and trisectors necessary to create a bunch of equal angles, then realized that many of those lines were not necessary once I figured out what claims to make. The problem actually seems pretty instructive in this regard.

Also, obligatory easy ≠ bad, and certainly nobody's going to coordinate bash this! xD

This post has been edited 1 time. Last edited by djmathman, Sep 22, 2020, 7:17 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

khina

993 posts

khina

#15 h Sep 22, 2020, 7:21 PM • 20 Y

Y by timon92, SnowPanda, Aryan-23, danepale, IAmTheHazard, Illuzion, OlympusHero, lneis1, samrocksnature, Corella, CrazyMathMan, 554183, centslordm, tigerzhang, IMUKAT, rayfish, pog, mathmax12, Mango247, aidan0626

Continuing off of dj, I thought this problem was actually quite difficult for a 1 (around IMO 2018 1), just because of how unbelievably strange the given condition was. I think it's hard to get started on the problem if you're not experienced with geometry - the key after all to solving this problem is to focus less on the condition and trying to find a way to construct the diagram accurately, which is why I think this is not an easy $1$ .

I also don't particularly like it, because of how contrived it was. Again, easy $\neq$ bad.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Smileyklaws

447 posts

Smileyklaws

#16 h Sep 22, 2020, 7:22 PM • 2 Y

Y by samrocksnature, Infinityfun

If no one can coordbash it, then it is a joy for some and a horror for the others.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Aryan-23

558 posts

Aryan-23

#17 h Sep 22, 2020, 7:22 PM • 2 Y

Y by A-Thought-Of-God, samrocksnature

Sol

This post has been edited 1 time. Last edited by Aryan-23, Sep 22, 2020, 9:50 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

HamstPan38825

8857 posts

HamstPan38825

#145 h Mar 10, 2024, 1:29 AM

Y by

Let $O$ be the circumcenter of triangle $APB$ . Then the quadrilaterals $CPOB$ and $AOPD$ are cyclic as $\angle POB + \angle PCB = 180^\circ$ given the angle condition. It follows that $\overline{DO}$ and $\overline{CO}$ bisect their respective angles, implying the result.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

AngeloChu

471 posts

AngeloChu

#146 h Apr 8, 2024, 12:40 PM

Y by

construct a circle around $ABP$
let it intersect $AD$ at $Q$ , and $BC$ at $R$
a simple angle chase yields that $DQP=2*QAP$ and $QAP=QPA$ , and $QPD=DQP$
therefore, $QDP$ is isosceles and since $QP$ is a chord, the angle bisector of $ADP$ is the perpendicular bisector of $QP$ and intersects this center of the circle $(ABP)$
similarly, the angle bisector of $BCP$ is the perpendicular bisector of $PR$ and intersects this center of the circle $(ABP)$
then, since $AB$ is a chord, it also passes through the center of the circle $(ABP)$

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Markas

105 posts

Markas

#147 h May 5, 2024, 12:56 PM

Y by

Let $\angle PAD = \alpha$ and $\angle PBC = \beta$ . Denote O as the circumcenter of $\triangle APB$ . Now $\angle AOB = 360 - 2\angle APB = 360 - 2(180 - 2\alpha - 2\beta) = 4\alpha + 4\beta$ $\Rightarrow$

$\angle OAB = \angle OBA = \frac{180 - 4\alpha - 4\beta}{2} = 90 - 2\alpha - 2\beta$ $\Rightarrow$ $\angle OAP = \angle OAB + \angle BAP = 90 - 2\alpha - 2\beta + 2\beta = 90 - 2\alpha$ . Now from AO = OP, we have that $\angle OAP = \angle OPA = 90 - 2\alpha$ $\Rightarrow$ $\angle OAD + \angle OPD = \angle OAP + \angle PAD + \angle OPA + \angle APD = 90 - 2\alpha + \alpha + 90 - 2\alpha + 3\alpha = 180^\circ$ $\Rightarrow$ AOPD is cyclic.

Since AOPD is cyclic we have that $\angle OAP = \angle ODP = 90 - 2\alpha$ . Since $\angle ADP = 180 - \angle DAP - \angle APD = 180 - 4\alpha$ , then $\angle ADO = 180 - 4\alpha - \angle ODP = 180 - 4\alpha - (90 - 2\alpha) = 90 - 2\alpha$ $\Rightarrow$ $\angle ADO = \angle ODP = 90 - 2\alpha$ $\Rightarrow$ DO is the angle bisector of $\angle ADP$ .

Similarly we will show BOPC is cyclic. From OP = OB, $\angle OPB = \angle OBP = \angle OBA + \angle ABP = 90 - 2\alpha - 2\beta + 2\alpha = 90 - 2\beta$ . Now $\angle OPC + \angle OBC = \angle OPB + \angle BPC + \angle OBP + \angle PBC = 90 - 2\beta + 3\beta + 90 - 2\beta + \beta = 180^\circ$ $\Rightarrow$ BOPC is cyclic $\Rightarrow$ $\angle OPB = \angle OCB = 90 - 2\beta$ . Also $\angle PCO = \angle PCB - \angle BCO = 180 - \angle CPB - \angle CBP - \angle BCO = 180 - 3\beta - \beta - (90 - 2\beta) = 90 - 2\beta$ $\Rightarrow$ $\angle PCO = \angle BCO = 90 - 2\beta$ $\Rightarrow$ CO is the angle bisector of $\angle PCB$ .

From this it follows that $DO \cap CO = O$ , where DO and CO are the angle bisectors of $\angle ADP$ and $\angle PCB$ , respectively. The last thing we need is, to show that the perpendicular bisector of AB passes trough O, which is obvious since O is the circumcenter of $\triangle APB$ $\Rightarrow$ the internal bisectors of angles $\angle ADP$ and $\angle PCB$ and the perpendicular bisector of segment AB meet in a point, which is explicitly O - the circumcenter of $\triangle APB$ $\Rightarrow$ we are ready.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

cj13609517288

1875 posts

cj13609517288

#148 h May 15, 2024, 3:08 PM

Y by

Let $\alpha=\angle DAP$ and $\beta=\angle CBP$ . Let $\omega=(ABP)$ meet $AD$ and $BC$ again at $Q_1$ and $Q_2$ , respectively. Then $\angle Q_1BP=\angle Q_1AP=\alpha$ , so $BQ_1$ bisects $\angle ABP$ . Thus $Q_1$ is the midpoint of arc $AP$ . Thus $\angle DQ_1P=2\alpha=\angle DPQ_1$ , so the angle bisector of $\angle PDA$ is just the perpendicular bisector of $Q_1P$ . Similarly, the angle bisector of $\angle PCB$ is just the perpendicular bisector of $Q_2P$ . These three lines clearly meet at the center of $\omega$ , so we are done. $\blacksquare$

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

shendrew7

792 posts

shendrew7

#149 h Jun 5, 2024, 4:18 PM

Y by

Consider the circumcenter $O$ of $\triangle ABP$ , which we claim is the desired concurrency point:

$O$ lies on the perpendicular bisector of $AB$ .
$O$ lies on $(BCP)$ , as $\measuredangle BOP = 2 \measuredangle BAP = \measuredangle CBP + \measuredangle BPC = \measuredangle BCP$ . Additionally, $BO = PO$ tells us $O$ is the midpoint of arc $BP$ , which means $O$ lies on the bisector of $\angle PCB$ .
$O$ lies on the bisector of $\angle ADP$ analogously. $\blacksquare$

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

peace09

5417 posts

peace09

#150 h Jun 21, 2024, 4:08 PM • 3 Y

Y by bjump, OronSH, LeonidasTheConquerer

One of the better things I've drawn in a while

(Edit: Oops I forgot CO below but it should be updated in the link!)

https://cdn.artofproblemsolving.com/attachments/4/d/d724117e3092b7f3b8c40fc5808d6064c5ed33.png

Denote the given angles by $\alpha,2\alpha,3\alpha$ and $\beta,2\beta,3\beta$ in the order they appear; and let the bisectors of $\angle PAB$ and $\angle PBA$ meet $BC$ and $AD$ at $E$ and $F$ , respectively. We have:

$\angle PAF=\angle PBF=\alpha\implies(PABF)$
$\angle FAP=\angle FPA=\alpha\implies FA=FP$
$\angle DFP=\angle DPF=2\alpha\implies DF=DP$

and

$\angle PBE=\angle PAE=\beta\implies(PABE)$
$\angle EBP=\angle EPB=\beta\implies EB=EP$
$\angle CEP=\angle CPE=2\beta\implies CE=CP$ .

Then, the given lines are the perpendicular bisectors of $AB$ , $EP$ , and $FP$ , which concur at the circumcenter of $(ABEFP)$ .

This post has been edited 1 time. Last edited by peace09, Jun 21, 2024, 4:11 PM

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

gracemoon124

872 posts

gracemoon124

#151 h Jul 4, 2024, 12:39 AM • 5 Y

Y by peace09, dolphinday, OronSH, bjump, LeonidasTheConquerer

solved with help from peace09 (: thanks for the problem rec <3

We bisect and trisect the angles that are 2x or 3x another. This is supposed to suggest another center -- the circumcenter of $\triangle ABP$ , which we'll name $O$ .

If we let $\angle PAD = \alpha$ and $\angle CBP = \beta$ , we have $\angle PDA = 180-4\alpha = 180 - 2\angle ABP = 180-\angle AOP.$ Similarly, $\angle BCP = 180 - \angle BOP$ . This means quadrilaterals $AOPD$ and $BOPC$ are cyclic.

Because $OA=OP$ , $OD$ is the angle bisector of $\angle ADP$ . Similarly, $OC$ is $\angle BCP$ 's bisector. Thus, the two angle bisectors intersect at $O$ , which lies on the perpendicular bisector of $AB$ . $\square$

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Mr.Sharkman

490 posts

Mr.Sharkman

#152 h Nov 7, 2024, 1:52 PM

Y by

Let $O$ be the circumcenter of $\triangle PAD.$ We claim that this works. Notice that $POAD$ and $POBC$ are cyclic. Thus, we are done.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Eka01

204 posts

Eka01

#153 h Nov 13, 2024, 5:35 PM

Y by

Let $O$ be the circumcenter of $\Delta ABP$ . Then easy angle chase gives us that $AOPD$ and $BOPC$ are cyclic. Due to fact $5$ , this means that $O$ passes through angle bisectors of $\angle PCB$ and $\angle ADP$ as desired.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

megahertz13

3177 posts

megahertz13

#154 h Nov 23, 2024, 7:57 PM

Y by

Let $O$ be the circumcenter of $\triangle{ABP}$ . We will show that these three lines all pass through $O$ .

First, $O$ clearly lies on the perpendicular bisector of segment $AB$ .

Claim: $BOPC$ and $AOPD$ are cyclic.

Note that $\angle{BCP}=180-\angle{CBP}-\angle{BPC}=180-2\cdot\angle{BAP}=180-\angle{BOP},$ so $BOPC$ is cyclic.

Similarly, $\angle{ADP}=180-\angle{PAD}-\angle{DPA}=180-2\angle{PBA}=180-\angle{AOP},$ so $AOPD$ is cyclic.

Since $O$ lies on the perpendicular bisector of $AP$ and $(DPA)$ , we have $\angle{ADO}=\angle{ODP}$ . Similarly, $\angle{PCO}=\angle{OCB}$ , done.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

SimplisticFormulas

84 posts

SimplisticFormulas

#155 h Dec 21, 2024, 3:27 PM

Y by

A classic example of a single construction tearing the question apart.

Let the given angle bisectors meet in $X$ . Construct $M$ , $N$ on $BC,AD$ with $BM=MP$ and $AN=NP$ . Let $O$ be the circumcentre of $\triangle ABP$ . Observe that
$\angle MBP=\angle MPB= \frac{1}{2}\angle BAP$ and $\angle CMP= \angle CPM$ so $M \in \odot(ABP)$ . Similarly, $N \in \odot(ABP)$ . Note that the angle bisectors of $\angle PCB$ and $\angle PDA$ are the perpendicular bisectors of $MP$ , $NP$ , which meet, by definition, in $O$ , which lies, by definition, on the perpendicular bisector of $AB$ .

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

CANTBANKAN

3 posts

CANTBANKAN

#156 h Jan 12, 2025, 8:34 PM

Y by

Let $O$ be the circumcenter of $\triangle ABP$ . Angle chasing implies that $\square OPBC$ and $\square OPAD$ are cyclic. The intersection of the internal bisectors of $\angle PDA$ and $\angle PCB$ is therefore $O$ , which obviously lies on the perpendicular bisector of $AB$ .

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Scilyse

386 posts

Scilyse

#157 h Jan 17, 2025, 10:51 AM

Y by

itslumi wrote:

What a disaster.

And here we see the convictions of someone who has no taste.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

Ilikeminecraft

312 posts

Ilikeminecraft

#158 h Mar 15, 2025, 1:52 AM

Y by

Let $O = (ADP)\cap(BPC).$ We get $\angle AOP = \angle AOD + \angle DOP = \angle APD + \angle DAP = 2\angle ABP.$ Thus, $O$ is circumcenter of $(ABP).$ Clearly, $\angle ODP = \angle OAP = \angle OPA = \angle ODA$ which finishes.

Z K Y

The post below has been deleted. Click to close.

This post has been deleted. Click here to see post.

endless_abyss

30 posts

endless_abyss

#159 h Mar 24, 2025, 5:24 PM

Y by

Nice! This is such a satisfying problem once you identify the intersection point!

Let $Q$ be the circumcentre of $B A P$ , we claim that

$D Q A P$ and $C Q B P$ are concyclic.

This follows from $\angle D A P = \angle D Q P$ and $\angle P B C = \angle P Q C$

Further, note that $\angle A Q P = 2 \angle A B P$
so, $\angle A Q D = 3x$ ,
$\angle Q A B = 90 - 2 x - 2 y$
so, $\angle A D Q = 90 - 2x = \angle Q D P$ and in triangle $C B P$ the angle chase follows similarly, so $Q$ is the desired concurrence point.

:starwars:

Z K Y